Re: [FFmpeg-devel] [PATCH] mxfdec: check edit_rate also for physical_track
On 15/11/15 18:07, Andreas Cadhalpun wrote: > Previously only the edit_rate of material_track was checked. > If it's negative, it causes assertion failures in av_rescale_rnd. > > Signed-off-by: Andreas Cadhalpun> --- > libavformat/mxfdec.c | 10 ++ > 1 file changed, 10 insertions(+) > > diff --git a/libavformat/mxfdec.c b/libavformat/mxfdec.c > index 99c8fed..429f46a 100644 > --- a/libavformat/mxfdec.c > +++ b/libavformat/mxfdec.c > @@ -1776,6 +1776,16 @@ static int > mxf_parse_physical_source_package(MXFContext *mxf, MXFTrack *source_t > continue; > } > > +if (physical_track->edit_rate.num <= 0 || > +physical_track->edit_rate.den <= 0) { > +av_log(mxf->fc, AV_LOG_WARNING, > + "Invalid edit rate (%d/%d) found on structural" > + " component #%d, defaulting to 25/1\n", > + physical_track->edit_rate.num, > + physical_track->edit_rate.den, i); > +physical_track->edit_rate = (AVRational){25, 1}; > +} > + > for (k = 0; k < > physical_track->sequence->structural_components_count; k++) { > if (!(mxf_tc = mxf_resolve_timecode_component(mxf, > _track->sequence->structural_components_refs[k]))) > continue; > LGTM, but I am curious as to where you have seen examples of negative edit rates. (I have found examples in mxf code where signed and unsigned have been muddled leading to invalid figures and I am wondering if this is another) -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/mxfenc: Only store user comment related tags when needed
On 10/11/15 12:28, Michael Niedermayer wrote: > From: Michael Niedermayer> > Also support disabling them as they seem to cause problems to some > Users. They are also not allowed in IRT D-10 thus the default for > mxf_d10 is not to write them > > This also decreases the filesize when no user comment are stored > > Signed-off-by: Michael Niedermayer > --- > doc/muxers.texi | 15 ++ > libavformat/mxfenc.c | 46 + > tests/ref/lavf/mxf | 12 > tests/ref/lavf/mxf_d10 |2 +- > tests/ref/lavf/mxf_opatom|2 +- > tests/ref/lavf/mxf_opatom_audio |4 +-- > tests/ref/seek/lavf-mxf | 44 +-- > tests/ref/seek/lavf-mxf_opatom_audio | 54 > +- > 8 files changed, 107 insertions(+), 72 deletions(-) > > [..] > @@ -2038,6 +2051,9 @@ static int mxf_write_header(AVFormatContext *s) > return -1; > } > > +if (!av_dict_get(s->metadata, "comment_", NULL, AV_DICT_IGNORE_SUFFIX)) > +mxf->store_user_comments = 0; > + > for (i = 0; i < s->nb_streams; i++) { > AVStream *st = s->streams[i]; > MXFStreamContext *sc = av_mallocz(sizeof(*sc)); > @@ -2650,12 +2666,14 @@ static int mxf_interleave(AVFormatContext *s, > AVPacket *out, AVPacket *pkt, int > { "smpte349m", "SMPTE 349M (1485 Mbps mappings)",\ >0, AV_OPT_TYPE_CONST, {.i64 = 6}, -1, 7, AV_OPT_FLAG_ENCODING_PARAM, > "signal_standard"},\ > { "smpte428", "SMPTE 428-1 DCDM",\ > - 0, AV_OPT_TYPE_CONST, {.i64 = 7}, -1, 7, AV_OPT_FLAG_ENCODING_PARAM, > "signal_standard"}, > + 0, AV_OPT_TYPE_CONST, {.i64 = 7}, -1, 7, AV_OPT_FLAG_ENCODING_PARAM, > "signal_standard"},\ Stray \ from previous patch attempt? > > > > static const AVOption mxf_options[] = { > MXF_COMMON_OPTIONS > +{ "store_user_comments", "", > + offsetof(MXFContext, store_user_comments), AV_OPT_TYPE_BOOL, {.i64 = > 1}, 0, 1, AV_OPT_FLAG_ENCODING_PARAM}, > { NULL }, > }; > > @@ -2670,6 +2688,8 @@ static const AVOption d10_options[] = { > { "d10_channelcount", "Force/set channelcount in generic sound essence > descriptor", >offsetof(MXFContext, channel_count), AV_OPT_TYPE_INT, {.i64 = -1}, -1, > 8, AV_OPT_FLAG_ENCODING_PARAM}, > MXF_COMMON_OPTIONS > +{ "store_user_comments", "", > + offsetof(MXFContext, store_user_comments), AV_OPT_TYPE_BOOL, {.i64 = > 0}, 0, 1, AV_OPT_FLAG_ENCODING_PARAM}, > { NULL }, > }; > > diff --git a/tests/ref/lavf/mxf b/tests/ref/lavf/mxf > [..] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/mxfenc: Only store user comment related tags when needed
On 09/11/15 17:56, Michael Niedermayer wrote: > From: Michael Niedermayer> > Also support disabling them as they seem to cause problems to some > Users. > > [..] > -for (i = 0; i < local_tag_number; i++) { > +for (i = 0; i < FF_ARRAY_ELEMS(mxf_local_tag_batch); i++) { > avio_wb16(pb, mxf_local_tag_batch[i].local_tag); > avio_write(pb, mxf_local_tag_batch[i].uid, 16); > } > +if (mxf->store_user_comments) > +for (i = 0; i < FF_ARRAY_ELEMS(mxf_local_tag_batch); i++) { Is this right? I expected:- for (i = 0; i < FF_ARRAY_ELEMS(mxf_user_comments_local_tag); i++) { > +avio_wb16(pb, mxf_user_comments_local_tag[i].local_tag); > +avio_write(pb, mxf_user_comments_local_tag[i].uid, 16); > +} > } -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 5/5] dnxhddec: replace some 0s by DNXHD_VARIABLE
On 02/10/15 20:00, Christophe Gisquet wrote: > A series of 0 in a table can be confusing, and the corresponding checks > strange, so using a macro instead of that magic is more readable. > --- > libavcodec/dnxhddata.c | 10 +- > libavcodec/dnxhddata.h | 3 +++ > libavcodec/dnxhddec.c | 6 -- > 3 files changed, 12 insertions(+), 7 deletions(-) > > [..] > +/** Indicate that a CIDEntry value must be read in the bitstream */ > +#define DNXHD_VARIABLE 0 > + I'm all for more readable, presumably atm its only ever 0 but may change in the future, in which case some idea of what it represents may be useful, I am sure the spec has more than one variable ;) > [..] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [RFC] avformat/mxfenc: stop encoding if unfilled video packet
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 04/10/15 13:07, Tomas Härdin wrote: > On Mon, 2015-09-28 at 15:11 +0200, Tobias Rapp wrote: >> On 19.09.2015 22:49, Tomas Härdin wrote: >>> On Wed, 2015-09-16 at 14:33 +0200, Tobias Rapp wrote: Hi, attached patch fixes ticket #4759 but I guess it is a bit too hasty to always abort transcoding if a single frame cannot be written. I gue ss it would be better to check for some "exit_on_error" like flag set but couldn't find out how to achieve that. Any comments would be appreciated. Regards, Tobias >>> >>> From 7d6f8de2a411817c970a19d8766e69b6eb604132 Mon Sep 17 00:00:00 2001 From: Tobias RappDate: Mon, 14 Sep 2015 12:06:22 +0200 Subject: [PATCH] avformat/mxfenc: stop encoding if unfilled video p acket occurs Fixes ticket #4759. --- libavformat/mxfenc.c | 14 +- 1 file changed, 9 insertions(+), 5 deletions(-) >> [...] >>> >>> Is this really better than not writing anything? >>> >>> /Tomas >> >> (Sorry for the long delay in answering, I was caught by a flu last we ek.) >> >> I want to transcode thousands of files and want to get some indicatio n >> from ffmpeg if the transcoding has been successful (all frames have b een >> transcoded) or not. Currently I am using the process exit code to ver ify >> that. >> >> There are two different user stories when using ffmpeg: >> a) "process the input data and exit with error as early as possible i n >> case of errors/problems" >> b) "process the input data and avoid exit with error as long as possi ble >> in case of error/problems" >> >> If I understand correctly I can basically switch between (a) and (b) >> mode by passing the "-xerror" option to ffmpeg or not. So my plan is to >> transcode all my files with "-xerror" and assume that >99% of the fil es >> will pass. The small amount of failing ones can then be analyzed for >> problems manually and possibly have a re-run without "-xerror". >> >> Unfortunately the "-xerror" option affects only ffmpeg but not the >> libraries, so I cannot use it do decide if "mxf_write_packet" should >> return with an error when the video packet size doesn't match the CBR >> constraints. > > For me the most important thing is that anything dealing with MXF shou ld > never write invalid files. +1 for sure. > I'm not sure whether the current code is > broken per se, but it does look suspicious. Perhaps someone else has a > better idea? > Truncate/pad mis sized frames to allow muxing to continue, and issue a warning (as at present)? > /Tomas > > > > ___ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > http://ffmpeg.org/mailman/listinfo/ffmpeg-devel > - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.22 (GNU/Linux) iQEcBAEBAgAGBQJWEiJtAAoJEAwL/ESLC/yDGQAIAJmqWRatBPd/2SXuzyFGWqB2 sg/FBpbzRqFGxr8dOfqfSkwLWu+EUL66UZKu5liVKkeksiUgx4sPtj6DFGFX7ozV g2DbJxvsFG18ES9C8OHc3qRDsgF4Z+F4GuScWMPrVcyvimSdHeajVkS8slrZlg2Y tMGQSin5jLwzv1pGIhCOdQLEndrr+PzwajPI837UBW2e7znWDb1uNHBy1yiPzcf+ 0pyZluyhkqW6IBzgO34Dc39DfQdGrtDWlzyUacT7nQz/4W5q2KAAhUhBTNtqTabK KCtSUfpqWgu6P2xxNvi87F9acqYprS80RC/ovrdgsXiXTNGSbYVscY748xg8mgA= =Clgr -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 5/5] dnxhddec: replace some 0s by DNXHD_VARIABLE
On 05/10/15 08:51, Christophe Gisquet wrote: > Hi, > > 2015-10-05 8:44 GMT+02:00 tim nicholson <nichot20-at-yahoo@ffmpeg.org>: >> On 02/10/15 20:00, Christophe Gisquet wrote: >>> A series of 0 in a table can be confusing, and the corresponding checks >>> strange, so using a macro instead of that magic is more readable. >>> --- >>> libavcodec/dnxhddata.c | 10 +- >>> libavcodec/dnxhddata.h | 3 +++ >>> libavcodec/dnxhddec.c | 6 -- >>> 3 files changed, 12 insertions(+), 7 deletions(-) >>> >>> [..] >>> +/** Indicate that a CIDEntry value must be read in the bitstream */ >>> +#define DNXHD_VARIABLE 0 >>> + >> >> I'm all for more readable, presumably atm its only ever 0 but may change >> in the future, in which case some idea of what it represents may be >> useful, I am sure the spec has more than one variable ;) > > Actually, it's a convenience value, that we put there after noticing > new samples. > > We don't have access to the DNxHR specs, but it seems a "profile" can > have differing bitdepths. I haven't yet verified, but the frame header > probably allows to distinguish them (DNXHD_VARIABLE applies to the HR > version). > That's fine, I just thought there ought to be a better name for it. Just calling it VARIABLE begs the question "what is variable?" If you are trying to make the code clearer, give some clue as to what it relates to. However if atm you aren't quite sure but think it might be handy later, then I can see that that becomes a bit difficult. I thought JJ had got hold of a copy of the specs... > I have, down the line, things that may need to split the HR versions. > -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avfilter/vf_tinterlace: add mergex2 mode
On 02/10/15 11:03, Paul B Mahol wrote: > On 10/1/15, Paul B Maholwrote: >> Signed-off-by: Paul B Mahol >> --- >> doc/filters.texi| 23 +++ >> libavfilter/tinterlace.h| 1 + >> libavfilter/vf_tinterlace.c | 17 - >> 3 files changed, 36 insertions(+), 5 deletions(-) > > > So is this ok now? This is supposed to replicate DoubleWeave from > avisynth/vapoursynth. No further comments from me, the desired result makes more sense, and I think the code is correct to achieve that, but I hoped others might check the logic too. > [..] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avfilter/vf_tinterlace: add mergex2 mode
On 30/09/15 14:39, Paul B Mahol wrote: > Signed-off-by: Paul B Mahol> --- > doc/filters.texi| 24 > libavfilter/tinterlace.h| 1 + > libavfilter/vf_tinterlace.c | 13 ++--- > 3 files changed, 35 insertions(+), 3 deletions(-) > > diff --git a/doc/filters.texi b/doc/filters.texi > index a4d828e..0a8588d 100644 > --- a/doc/filters.texi > +++ b/doc/filters.texi > @@ -10774,6 +10774,30 @@ Output: > 1 1 2 2 3 3 4 > @end example > > +@item mergex2, 7 > +Move odd frames into the upper field, even into the lower field, > +generating a double height frame at same frame rate. > +@example > + --> time > +Input: > +Frame 1 Frame 2 Frame 3 Frame 4 > + > +1 2 3 4 > +1 2 3 4 > +1 2 3 4 > +1 2 3 4 > + > +Output: > +1 2 3 4 > +2 3 4 5 > +1 2 3 4 > +2 3 4 5 > +1 2 3 4 > +2 3 4 5 > +1 2 3 4 > +2 3 4 5 > +@end example > + > I can see the usefulness of merging, but making any particular frame both the lower field in one merged frame, and upper in the next, sounds like a recipe for eye watering judder on the output. I think one would need to explain how would one then strip alternate frames to get a stream with a consistent field dominance(and how to select even or odd frames to get the right one). > @end table > > [..] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [DECISION] Server "move"
On 21/09/15 20:44, Carl Eugen Hoyos wrote: > Michael Niedermayer gmx.at> writes: > >> This "decision" thingy, is intended to confirm that >> * Stuff should be moved to the telepoint server, possibly duplicates >> should be left at/moved to the hetzner server for redundancy, if >> this is technically possible (maybe it should be called copy instead >> of move in that case) > > Copy sounds like a good idea to me. > > VDD15 again confirmed that nobody wants to do the work > necessary to move to videolan (which would also sound > ok to me). > Nobody, or nobody who was at VDD15? What work is necessary? I would +1 Lou's comments, some of us who might be able to help are still unsure of what happened at VDD15. AIUI we would like 2 servers, and having accepted one offer, a server has already been built and awaits commissioning. the second is still undecided. Could that not be on Videolan? What are the issues? Having accepted an offer which has led to those making that offer invest time, effort, and presumably money, in setting something up. I would imagine that they would be somewhat upset if we then turn round and say we have changed our mind. Especially without a clear alternative. It feels rather like we are going round in circles. > Carl Eugen > [..] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] mxfdec: set AVFMT_SEEK_TO_PTS demuxer flag
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 14/08/15 13:27, Tomas Härdin wrote: On Mon, 2015-08-10 at 10:14 +0200, Tomas Härdin wrote: On Sun, 2015-08-09 at 20:32 +0200, Marton Balint wrote: Since 53f2ef2c4afb1d49a679dea9163cb0e4671f3117 seeking is done using PTS. Signed-off-by: Marton Balint c...@passwd.hu --- libavformat/mxfdec.c | 1 + 1 file changed, 1 insertion(+) diff --git a/libavformat/mxfdec.c b/libavformat/mxfdec.c index 2d921db..5734976 100644 --- a/libavformat/mxfdec.c +++ b/libavformat/mxfdec.c @@ -3210,6 +3210,7 @@ static int mxf_read_seek(AVFormatContext *s, i nt stream_index, int64_t sample_ti AVInputFormat ff_mxf_demuxer = { .name = mxf, .long_name = NULL_IF_CONFIG_SMALL(MXF (Material eXchange Format)), +.flags = AVFMT_SEEK_TO_PTS, .priv_data_size = sizeof(MXFContext), .read_probe = mxf_probe, .read_header= mxf_read_header, Yeah, I seem to recall this when swearing at the seek code in mxfdec some years ago. I'll wait a few days to see if any other MXF guys hav e something to say here or on IRC, then I suppose I'll push Only the seek code you swear at? /Tomas Pushed. Hopefully everything worked alright Just back from leave, so time will tell /Tomas [..] - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.22 (GNU/Linux) iQEcBAEBAgAGBQJV3BIoAAoJEAwL/ESLC/yD4jgH/35vOcJkv8daPCvNiXc+p8HD DJc3YBFINl3wY8HohQJWcjYTfma8DVJS4XPCIMqsYpnzPvWH+VBp5hYxQ1UwrRDY trRjuMTtcfDp9kL3O1cGKHPDV0eyJXvoyR8gw1zpKdZsVE4In7VnQRLahj//2HFt +tWcuPG+IhMrgyOMM5UH5rSd5doCQgkLC3iouX40NTmpzo5fGAdqH8+4qrMJSY6e Uth3xK7hHRQqnbDEYNmKMOSeVasnwaxx26y2hCtVUaUypnyhH7lkY2ndfURg0pD3 80yCPI253cy1tbtsmRyph3z/BXTsa3xPUFIcdcKjpnQuv7pio/a+w3a1UqpE8PI= =1mAZ -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] FFmpegs future and resigning as leader
On 03/08/15 20:43, Michael Niedermayer wrote: On Mon, Aug 03, 2015 at 08:51:10AM +0100, tim nicholson wrote: On 31/07/15 17:19, Michael Niedermayer wrote: On Fri, Jul 31, 2015 at 05:37:12PM +0200, Clément Bœsch wrote: [...] So in order for the community to continue this, I'd say we probably need to have some help for: - guidelines on the merge strategies - step-by-step on the release process - some overview on the sysadmin state (like, what happens with the recent server offers?) i accepted the 2 offers which noone objected against (on the ML), so FFmpeg should get 2 boxes that things can be moved to and which should be dependable it does need volunteers doing the work. Maybe lou, beastd, tim nich, roberto and you would be willing to help move things to them I am up for assisting where I can, but August is going to be mainly holiday for me ;) it was discussed to move things into virtual machines (qemu) for higher security, isolation and ease of future moving. If we go kvm etc, prebuilding a base image beforehand that can then be forked for postfix/Apache/trak etc would be useful. yes In fact there are some pre-built images already for some things, e.g https://www.turnkeylinux.org/issue-tracking?page=1 has these been build by someone who knows about security ? or would it require a security audit to ensure no auth stuff /session keys / host keys / prng state / whatever is carried over from the public image ? From a quick look they seem to know what they are doing. On first run it behaves somewhat like the end of an install process first boot, so each instance is different with regenerated keys. When I tried 2 separate instances from scrach I confirmed they had different host keys. I didn't go deeper than that at the time. Obviously prng integrity is important and would need checking. [...] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] FFmpegs future and resigning as leader
On 31/07/15 17:19, Michael Niedermayer wrote: On Fri, Jul 31, 2015 at 05:37:12PM +0200, Clément Bœsch wrote: [...] So in order for the community to continue this, I'd say we probably need to have some help for: - guidelines on the merge strategies - step-by-step on the release process - some overview on the sysadmin state (like, what happens with the recent server offers?) i accepted the 2 offers which noone objected against (on the ML), so FFmpeg should get 2 boxes that things can be moved to and which should be dependable it does need volunteers doing the work. Maybe lou, beastd, tim nich, roberto and you would be willing to help move things to them I am up for assisting where I can, but August is going to be mainly holiday for me ;) it was discussed to move things into virtual machines (qemu) for higher security, isolation and ease of future moving. If we go kvm etc, prebuilding a base image beforehand that can then be forked for postfix/Apache/trak etc would be useful. In fact there are some pre-built images already for some things, e.g https://www.turnkeylinux.org/issue-tracking?page=1 that would just need the data migrating. moving to the new servers could be done before or after moving to VMs of course or while doing it in the same step [...] ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] Add support for j2ki/mxf
On 23/07/15 11:32, Matthieu Bouron wrote: On Thu, Jul 23, 2015 at 11:04 AM, tim nicholson nichot20-at-yahoo@ffmpeg.org wrote: On 22/07/15 14:29, Matthieu Bouron wrote: Hello there, The following patchset add support for j2ki/mxf files (which store frames using the separate fields / segmented frame layouts, ie: fields are stored in seperate frames) through the insertion of filters at the ffmpeg/ffplay level. It differs from my original attempt which introduced modifications at the decoder level, which in my opinion was wrong and introduced several limitations, such as: [..] I may be missing something here, as my mind has been on other things. But currently most mxf material I see that ffmpeg correctly handles is frame wrapped with the frame layout as type 1 (SEPARATE_FIELDS). Can you provide samples with this layout ? I might have misinterpreted the MXF specs on this one. My understating is that, separated fields (as well as segmented frame) means that the fields are stored in a different packet (that might no be the right world for it). Also it might be specific to jpeg2000 (see below), i'll have to look again at the mxf/jpeg2000 spec again. I have added a 4 frame example to the trac ticket of interlaced IMX50 material. according to mxf2raw (from bmxlib) this file is flagged SEPARATE_FIELDS. [..] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] Add support for j2ki/mxf
On 22/07/15 14:29, Matthieu Bouron wrote: Hello there, The following patchset add support for j2ki/mxf files (which store frames using the separate fields / segmented frame layouts, ie: fields are stored in seperate frames) through the insertion of filters at the ffmpeg/ffplay level. It differs from my original attempt which introduced modifications at the decoder level, which in my opinion was wrong and introduced several limitations, such as: [..] I may be missing something here, as my mind has been on other things. But currently most mxf material I see that ffmpeg correctly handles is frame wrapped with the frame layout as type 1 (SEPARATE_FIELDS). AFAIK these do not use filter insertion etc to work. Are these therefore handled at the decoder level, in which case:- a: they will not benefit from doing things at the format level. b: J2K will be an exception to the current way of doing things, which could add confusion. If they are not currently handled at the decoder level, the surely the format level must already handle this material correctly so why is your patch needed in this form? -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH]lavf/mxf: Map codec_tag for Avid files if everything else fails
On 22/07/15 11:10, Carl Eugen Hoyos wrote: Tomas Härdin tomas.hardin at codemill.se writes: New patch attached. Looks alright. You can reindent in a separate patch if you like. Merged by Michael with the reindentation. There is still a cropping issue for the given sample: Am I correct that the generic descriptors 0x3208 and 0x3209 should be used as display size with 0x320B as vertical offset (and 0x320A as horizontal offset)? That is my understanding. [..] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/mxfdec: Change version byte for JPEG2000 to match mxf.c and RP224v12
On 15/06/15 13:13, Michael Niedermayer wrote: This should make no difference as the byte is ignored Found-by: tim nicholson nicho...@yahoo.com Signed-off-by: Michael Niedermayer michae...@gmx.at --- libavformat/mxfdec.c |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libavformat/mxfdec.c b/libavformat/mxfdec.c index 7389555..b3c25b7 100644 --- a/libavformat/mxfdec.c +++ b/libavformat/mxfdec.c @@ -1156,7 +1156,7 @@ static const MXFCodecUL mxf_data_essence_container_uls[] = { }; static const MXFCodecUL mxf_codec_uls[] = { -{ { 0x06,0x0E,0x2B,0x34,0x04,0x01,0x01,0x09,0x04,0x01,0x02,0x02,0x03,0x01,0x01,0x00 }, 14, AV_CODEC_ID_JPEG2000 }, +{ { 0x06,0x0E,0x2B,0x34,0x04,0x01,0x01,0x07,0x04,0x01,0x02,0x02,0x03,0x01,0x01,0x00 }, 14, AV_CODEC_ID_JPEG2000 }, { { 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00 }, 0, AV_CODEC_ID_NONE }, }; LGTM ;) Also just noticed that we now have both MXFCodecUL ff_mxf_codec_uls[] and MXFCodecUL mxf_codec_uls[] in the mxf codebase. similar names similar functionality and one a subset of the other. Yet more UL confusion in mxf land to eventually untangle. -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/mxfdec: Detect jpeg2000 through codec_ul too
On 12/06/15 18:05, Michael Niedermayer wrote: Fixes Ticket2345 Signed-off-by: Michael Niedermayer michae...@gmx.at --- libavformat/mxfdec.c | 14 ++ 1 file changed, 14 insertions(+) diff --git a/libavformat/mxfdec.c b/libavformat/mxfdec.c index 78e2393..7389555 100644 --- a/libavformat/mxfdec.c +++ b/libavformat/mxfdec.c @@ -166,6 +166,7 @@ typedef struct MXFDescriptor { enum MXFMetadataSetType type; UID essence_container_ul; UID essence_codec_ul; +UID codec_ul; AVRational sample_rate; AVRational aspect_ratio; int width; @@ -974,6 +975,9 @@ static int mxf_read_generic_descriptor(void *arg, AVIOContext *pb, int tag, int case 0x3004: avio_read(pb, descriptor-essence_container_ul, 16); break; +case 0x3005: +avio_read(pb, descriptor-codec_ul, 16); +break; case 0x3006: descriptor-linked_track_id = avio_rb32(pb); break; @@ -1151,6 +1155,11 @@ static const MXFCodecUL mxf_data_essence_container_uls[] = { { { 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00 }, 0, AV_CODEC_ID_NONE }, }; +static const MXFCodecUL mxf_codec_uls[] = { +{ { 0x06,0x0E,0x2B,0x34,0x04,0x01,0x01,0x09,0x04,0x01,0x02,0x02,0x03,0x01,0x01,0x00 }, 14, AV_CODEC_ID_JPEG2000 }, _/ mxf.c already has:- { { 0x06,0x0E,0x2B,0x34,0x04,0x01,0x01,0x07,0x04,0x01,0x02,0x02,0x03,0x01,0x01,0x00 }, 14, AV_CODEC_ID_JPEG2000 }, /* JPEG2000 Codestream */ / and my copy of RP224v12 agrees with that registry version byte. Since the version byte is ignored this should make no functional difference, but it is inconsistent with what we already have, and the published spec. +{ { 0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00 }, 0, AV_CODEC_ID_NONE }, +}; + static const char* const mxf_data_essence_descriptor[] = { vbi_vanc_smpte_436M, }; @@ -1950,6 +1959,11 @@ static int mxf_parse_structural_metadata(MXFContext *mxf) /* TODO: drop PictureEssenceCoding and SoundEssenceCompression, only check EssenceContainer */ codec_ul = mxf_get_codec_ul(ff_mxf_codec_uls, descriptor-essence_codec_ul); st-codec-codec_id = (enum AVCodecID)codec_ul-id; +if (st-codec-codec_id == AV_CODEC_ID_NONE) { +codec_ul = mxf_get_codec_ul(mxf_codec_uls, descriptor-codec_ul); +st-codec-codec_id = (enum AVCodecID)codec_ul-id; +} + av_log(mxf-fc, AV_LOG_VERBOSE, %s: Universal Label: , avcodec_get_name(st-codec-codec_id)); for (k = 0; k 16; k++) { -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/mxfenc: Allow overriding /manual setting of the signal standard
On 06/06/15 02:23, Michael Niedermayer wrote: Signed-off-by: Michael Niedermayer michae...@gmx.at --- libavformat/mxfenc.c | 10 ++ 1 file changed, 10 insertions(+) diff --git a/libavformat/mxfenc.c b/libavformat/mxfenc.c index c612747..f2a7f0a 100644 --- a/libavformat/mxfenc.c +++ b/libavformat/mxfenc.c @@ -316,6 +316,7 @@ typedef struct MXFContext { uint32_t instance_number; uint8_t umid[16];/// unique material identifier int channel_count; +int signal_standard; uint32_t tagged_value_count; AVRational audio_edit_rate; } MXFContext; @@ -2104,6 +2105,8 @@ static int mxf_write_header(AVFormatContext *s) sc-signal_standard = 1; } +if (mxf-signal_standard = 0) +sc-signal_standard = mxf-signal_standard; } else if (st-codec-codec_type == AVMEDIA_TYPE_AUDIO) { if (st-codec-sample_rate != 48000) { av_log(s, AV_LOG_ERROR, only 48khz is implemented\n); @@ -2627,7 +2630,12 @@ static int mxf_interleave(AVFormatContext *s, AVPacket *out, AVPacket *pkt, int mxf_interleave_get_packet, mxf_compare_timestamps); } +#define COMMON_OPTIONS \ nit MXF_COMMON_OPTIONS ? feels more comfortable to me when grepping the code etc. +{ signal_standard, Force/set Sigal Standard,\ + offsetof(MXFContext, signal_standard), AV_OPT_TYPE_INT, {.i64 = -1}, -1, 8, AV_OPT_FLAG_ENCODING_PARAM}, + Isn't 07h the largest valid value (G.2.3)? Or do I misunderstand the constraint values? (AVOPTION not my strong point) I thought numeric options were deprecated these days in favour of more understandable string values, or are the string values also too obscure to be useful in this case? (Just wondering what the current protocol was on this nowadays, not a real concern in this case) static const AVOption mxf_options[] = { +COMMON_OPTIONS { NULL }, }; @@ -2641,6 +2649,7 @@ static const AVClass mxf_muxer_class = { static const AVOption d10_options[] = { { d10_channelcount, Force/set channelcount in generic sound essence descriptor, offsetof(MXFContext, channel_count), AV_OPT_TYPE_INT, {.i64 = -1}, -1, 8, AV_OPT_FLAG_ENCODING_PARAM}, +COMMON_OPTIONS { NULL }, }; @@ -2654,6 +2663,7 @@ static const AVClass mxf_d10_muxer_class = { static const AVOption opatom_options[] = { { mxf_audio_edit_rate, Audio edit rate for timecode, offsetof(MXFContext, audio_edit_rate), AV_OPT_TYPE_RATIONAL, {.dbl=25}, 0, INT_MAX, AV_OPT_FLAG_ENCODING_PARAM }, +COMMON_OPTIONS { NULL }, }; otherwise LGTM. -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] Test-please ignore
Test of address munging on new server -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/mxfenc: Accept MXF D-10 with 49.999840 Mbit/sec
On 01/06/15 20:35, Michael Niedermayer wrote: This is the maximum rate possible based on the frame size limit of MXF D-10 Signed-off-by: Michael Niedermayer michae...@gmx.at --- libavformat/mxfenc.c |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libavformat/mxfenc.c b/libavformat/mxfenc.c index e0ae14e..7e41c5c 100644 --- a/libavformat/mxfenc.c +++ b/libavformat/mxfenc.c @@ -2080,7 +2080,7 @@ static int mxf_write_header(AVFormatContext *s) sc-video_bit_rate = st-codec-bit_rate ? st-codec-bit_rate : st-codec-rc_max_rate; if (s-oformat == ff_mxf_d10_muxer) { -if (sc-video_bit_rate == 5000) { +if (sc-video_bit_rate = 4840 sc-video_bit_rate = 5000) { As well as allowing setting of correct rate for 3000/1001, this could allow setting of non standard bit rates between/including the two figures for both frame rates, which might happen if someone made a typo/guess. Surely better to do:- if (sc-video_bit_rate == 5000) (mxf-time_base.den == 25) sc-index = 3; else if (sc-video_bit_rate = 4840 (mxf-time_base.den != 25) sc-index = 5; else if ... if (mxf-time_base.den == 25) sc-index = 3; else sc-index = 5; } else if (sc-video_bit_rate == 4000) { -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/mxfenc: Correct klv_fill_key
On 29/05/15 23:26, Michael Niedermayer wrote: See SMPTE 377-1-2009 6.3.3 KLV Fill Items Signed-off-by: Michael Niedermayer michae...@gmx.at --- libavformat/mxfenc.c|2 +- tests/ref/lavf/mxf |6 +++--- tests/ref/lavf/mxf_d10 |2 +- tests/ref/lavf/mxf_opatom |2 +- tests/ref/lavf/mxf_opatom_audio |2 +- 5 files changed, 7 insertions(+), 7 deletions(-) diff --git a/libavformat/mxfenc.c b/libavformat/mxfenc.c index 63471e6..e0ae14e 100644 --- a/libavformat/mxfenc.c +++ b/libavformat/mxfenc.c @@ -334,7 +334,7 @@ static const uint8_t index_table_segment_key[] = { 0x06,0x0E,0x2B,0x34,0x02, static const uint8_t random_index_pack_key[] = { 0x06,0x0E,0x2B,0x34,0x02,0x05,0x01,0x01,0x0D,0x01,0x02,0x01,0x01,0x11,0x01,0x00 }; static const uint8_t header_open_partition_key[] = { 0x06,0x0E,0x2B,0x34,0x02,0x05,0x01,0x01,0x0D,0x01,0x02,0x01,0x01,0x02,0x01,0x00 }; // OpenIncomplete static const uint8_t header_closed_partition_key[] = { 0x06,0x0E,0x2B,0x34,0x02,0x05,0x01,0x01,0x0D,0x01,0x02,0x01,0x01,0x02,0x04,0x00 }; // ClosedComplete -static const uint8_t klv_fill_key[]= { 0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x01,0x03,0x01,0x02,0x10,0x01,0x00,0x00,0x00 }; +static const uint8_t klv_fill_key[]= { 0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x02,0x03,0x01,0x02,0x10,0x01,0x00,0x00,0x00 }; static const uint8_t body_partition_key[] = { 0x06,0x0E,0x2B,0x34,0x02,0x05,0x01,0x01,0x0D,0x01,0x02,0x01,0x01,0x03,0x04,0x00 }; // ClosedComplete I agree the registry version was incorrect, I think there are others too. So LGTM. -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/mxfenc: Correct klv_fill_key
On 01/06/15 15:01, Michael Niedermayer wrote: On Mon, Jun 01, 2015 at 07:11:38AM +0100, tim nicholson wrote: On 29/05/15 23:26, Michael Niedermayer wrote: See SMPTE 377-1-2009 6.3.3 KLV Fill Items Signed-off-by: Michael Niedermayer michae...@gmx.at --- libavformat/mxfenc.c|2 +- tests/ref/lavf/mxf |6 +++--- tests/ref/lavf/mxf_d10 |2 +- tests/ref/lavf/mxf_opatom |2 +- tests/ref/lavf/mxf_opatom_audio |2 +- 5 files changed, 7 insertions(+), 7 deletions(-) diff --git a/libavformat/mxfenc.c b/libavformat/mxfenc.c index 63471e6..e0ae14e 100644 --- a/libavformat/mxfenc.c +++ b/libavformat/mxfenc.c @@ -334,7 +334,7 @@ static const uint8_t index_table_segment_key[] = { 0x06,0x0E,0x2B,0x34,0x02, static const uint8_t random_index_pack_key[] = { 0x06,0x0E,0x2B,0x34,0x02,0x05,0x01,0x01,0x0D,0x01,0x02,0x01,0x01,0x11,0x01,0x00 }; static const uint8_t header_open_partition_key[] = { 0x06,0x0E,0x2B,0x34,0x02,0x05,0x01,0x01,0x0D,0x01,0x02,0x01,0x01,0x02,0x01,0x00 }; // OpenIncomplete static const uint8_t header_closed_partition_key[] = { 0x06,0x0E,0x2B,0x34,0x02,0x05,0x01,0x01,0x0D,0x01,0x02,0x01,0x01,0x02,0x04,0x00 }; // ClosedComplete -static const uint8_t klv_fill_key[]= { 0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x01,0x03,0x01,0x02,0x10,0x01,0x00,0x00,0x00 }; +static const uint8_t klv_fill_key[]= { 0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x02,0x03,0x01,0x02,0x10,0x01,0x00,0x00,0x00 }; static const uint8_t body_partition_key[] = { 0x06,0x0E,0x2B,0x34,0x02,0x05,0x01,0x01,0x0D,0x01,0x02,0x01,0x01,0x03,0x04,0x00 }; // ClosedComplete I agree the registry version was incorrect, applied I think there are others too. So LGTM. do you know of any specific ones ? or just a general feeling that theres more? I think I spotted a couple of others when I was trawling through, but I wanted to simplify the UL layout first to make it easier to spot these anomalies. I had in mind to do something along the lines of bmxlib using macros to define the common first 7 byte parts according to Registry where they are defined, then it might be easier to see some of these oddities also the MPEG video Descriptor short thingies (0x800X) differ from what ive seen in other implementations but i couldnt find a proper reference that lists them, i wonder if someone maybe knows where to look? Not sure either mxf spec only uses below 7F.FF! There are also errors in the normative table Annex H in S377, for example local tag 08.01 is listed twice with 2 different UL's, hence tidying things up is a bit of a 'mare. -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/mxfenc: Support storing signal standard
On 24/05/15 02:15, Michael Niedermayer wrote: also store 1 for D10 Signed-off-by: Michael Niedermayer michae...@gmx.at --- libavformat/mxfenc.c| 11 +++ tests/ref/lavf/mxf |6 +++--- tests/ref/lavf/mxf_d10 |2 +- tests/ref/lavf/mxf_opatom |2 +- tests/ref/lavf/mxf_opatom_audio |2 +- 5 files changed, 17 insertions(+), 6 deletions(-) Do we want to make this an edge case for D10 muxing only, or should we be aiming for it to be more universal? Personally I try to avoid edge cases but I can see that trying to be more universal could be difficult for a number of reasons in this case. If we keep this as an edge case perhaps the commit message should say so:- Support storing signal standard for D10 muxing or some such. (I can see a number of similar issues on the horizon too.) -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/mxfenc: Add color siting element
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 19/05/15 16:50, Michael Niedermayer wrote: On Tue, May 19, 2015 at 03:35:42PM +0100, tim nicholson wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 19/05/15 14:11, Michael Niedermayer wrote: On Tue, May 19, 2015 at 12:07:24PM +0100, tim nicholson wrote: On 19/05/15 01:33, Michael Niedermayer wrote: [...] the avchroma locations are explained by the ascii art above the enum : * X X 3 4 X X are luma samples, * 1 21-6 are possible chroma positions * X X 5 6 X 0 is undefined/unknown position and from this the english directions are exactly where the chroma samples are located, this also matches how H264 defines chroma locations. I cannot get the ascii art to make sense to me at all, it just doesn' t click. I do no see how you can represent co-siting in ascii art wit h a single character and the 3 position certainly doesn't do it for me! ascii art improved does it make sense now ? Better, but I still cannot see the difference between say 3 and 4, or 1 and 2. [...] My understanding of S377 using 2/3 ascii chars/pixel is:- 00h:- YC Y- YC -Y- YC Y- YC -Y- YC Y- YC -Y- 01h:- Y- C- Y- C- Y- Y- C- Y- C- Y- Y- C- Y- C- Y- 03h:- Y-Y- - -- C- -- Y-Y- 05h:- YCr Y- YCr YCb Y- YCb YCr Y- YCr 06h:- Y- Y- Y- Y- C- C- C- C- Y- Y- Y- Y- Not all samples are retained dependening upon the subsampling scheme. - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.22 (GNU/Linux) iQEcBAEBAgAGBQJVXYdQAAoJEAwL/ESLC/yD0NMH/jTJYQG8ZLtwMKbDWAfvkUzk FWhWJxVQl3JBftxmjkB7JRtRebX6XbDklUi2OQplIzAhbqLdQIJHgDC8enYRIRBf FJE4qTPltyJ/taOuXeM1IGwrAdbXql+wIakQbwXYfu07dZ3fLAFk63u6mBx+pLeo l3JpVwOMQsxvjnnrrEr73mKlVk9vo99+tjJiBng5UR02SLfp/6Blvyl4HhyCMYZr WvZf/pKhETmxjhYMhM/tBncgzeA5gW6q7onFsdBM8g+n/nxSF/nQrXXlkWlWN4hN M4gnWwVddKbdmaP/kZBzJsIfKnEsH7em2rFOseMrmtvSImDZVpIpRFkG8pVe1JY= =9Pkq -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/mxfenc: Add color siting element
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 21/05/15 11:47, Michael Niedermayer wrote: On Thu, May 21, 2015 at 08:20:48AM +0100, tim nicholson wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 19/05/15 16:50, Michael Niedermayer wrote: On Tue, May 19, 2015 at 03:35:42PM +0100, tim nicholson wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 19/05/15 14:11, Michael Niedermayer wrote: On Tue, May 19, 2015 at 12:07:24PM +0100, tim nicholson wrote: On 19/05/15 01:33, Michael Niedermayer wrote: [...] the avchroma locations are explained by the ascii art above the en um : * X X 3 4 X X are luma samples, * 1 21-6 are possible chroma positions * X X 5 6 X 0 is undefined/unknown position and from this the english directions are exactly where the chroma samples are located, this also matches how H264 defines chroma locations. I cannot get the ascii art to make sense to me at all, it just does n' t click. I do no see how you can represent co-siting in ascii art w it h a single character and the 3 position certainly doesn't do it for m e! ascii art improved does it make sense now ? Better, but I still cannot see the difference between say 3 and 4, or 1 and 2. Ive tried to improve the text and ASCII even more does it make sense now ? Still don't understand why two colour sample sites per pixel. Please don't waste any more time fiddling with this, I expect it will suddenly click in my brain when I least expect it. Oh hang on I think it just did, you superimpose the second art on top of the first so the 3 is on top of the X but the 4 is between them etc etc In which case I think I agree with your mappings, sorry its taken so lon g ;( [...] ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.22 (GNU/Linux) iQEcBAEBAgAGBQJVXeXUAAoJEAwL/ESLC/yD/XwH+wfNWVEQ877APxfpO4fGmkc9 4WDvc8E4tzwVda6FKw+LuSLFHAQcHnuaZLcwXRfLeXrQpgsZUEqagJ/A4A8eDsFp GI0BZLiHROOnHBAZ5yYNd3q6Y8lJump4LK/XgjOIURlnk5wEJ6nkM3sjqoP4JoDU EkEDf1ZYAeddRnu/hfZp7o8wNqhXoreh8FBj2Sznli4jDr680WfffwavNfJ6WSF7 QRfCZ9wNSw7JrVs02BsT90VDLXh45+2DeFfQqf2XC7apYZ/1gRY97hLMu+M52NQZ mgXTSR6gTiBU7EPwR7Mq5MS1x4Za8d2uPJy+OIdGOp4wUllrCMrgEnA+Nc47Hb0= =wsE6 -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/mxfenc: Add color siting element
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 21/05/15 11:29, Michael Niedermayer wrote: On Thu, May 21, 2015 at 08:20:48AM +0100, tim nicholson wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 19/05/15 16:50, Michael Niedermayer wrote: On Tue, May 19, 2015 at 03:35:42PM +0100, tim nicholson wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 19/05/15 14:11, Michael Niedermayer wrote: On Tue, May 19, 2015 at 12:07:24PM +0100, tim nicholson wrote: On 19/05/15 01:33, Michael Niedermayer wrote: [...] the avchroma locations are explained by the ascii art above the en um : * X X 3 4 X X are luma samples, * 1 21-6 are possible chroma positions * X X 5 6 X 0 is undefined/unknown position and from this the english directions are exactly where the chroma samples are located, this also matches how H264 defines chroma locations. I cannot get the ascii art to make sense to me at all, it just does n' t click. I do no see how you can represent co-siting in ascii art w it h a single character and the 3 position certainly doesn't do it for m e! ascii art improved does it make sense now ? Better, but I still cannot see the difference between say 3 and 4, or 1 and 2. [...] My understanding of S377 using 2/3 ascii chars/pixel is:- 00h:- YC Y- YC -Y- YC Y- YC -Y- YC Y- YC -Y- S377-2009 speaks about the first sample of the image only The first luma sample of the image is co-sited with the first color d ifference sample(s), as in ITU-R Rec 601, SMPTE 314M 4:1:1 or MPEG-2 4:2:2. ...and ITU-R Rec 601 dicates Orthogonal, line, field and frame repetitive. CR and CB samples co-sited with odd (1st, 3rd, 5th, etc.) Y samples in each line Your ASCII art matches my interpretation of this though 00h can be more than 4:2:2, it explicitly also lists 4:1:1 agreed. 01h:- Y- C- Y- C- Y- Y- C- Y- C- Y- Y- C- Y- C- Y- This implies 4:4:4 I was indicating sample sites, which I then qualified with a:- Not all samples are retained dependening upon the subsampling scheme. Y- C- Y- Y- C- Y- Y- C- Y- Y- C- Y- Y- C- Y- Y- C- Y- would be possible too not to mention the mxf text for 01h The color sample is sited at the point horizontally midway between th e luma sample on each line. does not say which luma samples are used for the midpoint also it speaks about color sample while 00h speaks about color difference sample(s) color samples would be RGB, difference Cb Cr. the MXF spec isnt really wording this very unambigously 03h:- Y-Y- - -- C- -- Y-Y- This ascii art seems to be faulty in some form it looks like chroma had more positions than luma horizontally Seems to have picked up an extra dash from what I wrote... 1 c to 2 Y horizontally and vertically 05h:- YCr Y- YCr YCb Y- YCb YCr Y- YCr 06h:- Y- Y- Y- Y- C- C- C- C- Y- Y- Y- Y- The text from the spec is The color sample is sited at the point vertically midway between the luma sample on each column, as in MPEG-2 4:2:0. i think your ascii art matches how i would interpet it for 4:4:0 or 4:4:4, not sure if you intended this to be not 4:2:0 I refer to my previous qualification:- Not all samples are retained dependening upon the subsampling scheme. Which matches how S413m tries to explain it by showing all sample sites even though some are discarded. Not withstanding any of this don't let it hold up the patch, I twas not my intention to bikeshed just trying to get my head around whats right. [...] - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.22 (GNU/Linux) iQEcBAEBAgAGBQJVXeJhAAoJEAwL/ESLC/yDLXAIALtpwEEHV/FdM+ClziRTjSEu pVL5pIMvBnR1V/QpXv6n00gFcGdYUBWOTH2HrKgcW+3gY64ofyhukm0k4vTaV5HE zoSbne33QKaoN/x9LSiiCLPn5oMmtOLQornhLLUxHp8aXMGl66ejlUJzJSQTfxHL ME4tV1+bZz7HOJlfFOWPI9+2DNI3eroRqguNHr4S8HMVY3NhH8ysuCueOcqLNoAQ zWGI/tPsuirTrvRt0iGC/iJhoh6PknVnndYoiMlYOzZs3/0294i9V3qEaQMgLmmr 9dD0XHUkTdo0Y7CP2zZZYDwz6KfWhP7sjmwZ6eIyI7wUkthsVKBFLD+x1tfIU/M= =KFhi -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/mxfenc: Add color siting element
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 21/05/15 11:59, Michael Niedermayer wrote: On Thu, May 21, 2015 at 09:00:56AM +0200, Tomas Härdin wrote: [..] Why is this guessing code in mxfenc? This should be something that' s taken care of before calling any muxer (like in avformat_write_header()), so everyone can benefit from it and so ther e's less risk of duplication when it's needed elsewhere. I've seen stunts like this pulled in more places than mxfenc, where muxers will do dubious things like parse certain kinds of essence. I am not a fan. nor am i, ill try to move it into common code +1 to both your comments its why I have hesitated on some of the other mxf metadata thats been requested. thanks [...] - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.22 (GNU/Linux) iQEcBAEBAgAGBQJVXeZTAAoJEAwL/ESLC/yDWeQIALJn7YieWo7yKqmyBjsHAA8p cSqYgLOcFufG7Hx1/sag8AfSExRe1LIcaQUR4RY8s54bUP4gmNk51je49QVlCAUo 3z6O0/0J/mkP/+M5GZ4fVKrKXj+HmL6t5keFbtxZnWFNz9nCjcWTNCGvMj0/KKfY G5nw/owzdu2intMdqgQYM0sblCmErAVogyasABn/lGqvWrt2H0KBs+eBICtiY1ac y667sBE4MrxCRtY3a1erDmXJSlOY7L5qFJFaYDLKx5i3mO4oeqZgHGrTnF7cojQ1 Q6iM23Zc220IIwyoCaEULGxenTS+FAz178wdAwO4Ydor238aDzlMRJdzi2iydYo= =+djy -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/mxfenc: Add color siting element
On 19/05/15 01:33, Michael Niedermayer wrote: The default is assumed to be 0xFF, which is what the 2009 spec lists, the older version i have lists 0 as the default. Signed-off-by: Michael Niedermayer michae...@gmx.at --- libavformat/mxfenc.c| 28 + tests/ref/lavf/mxf | 12 +-- tests/ref/lavf/mxf_d10 |2 +- tests/ref/lavf/mxf_opatom |2 +- tests/ref/lavf/mxf_opatom_audio |2 +- tests/ref/seek/lavf-mxf | 44 +++ 6 files changed, 59 insertions(+), 31 deletions(-) diff --git a/libavformat/mxfenc.c b/libavformat/mxfenc.c index 659c34f..0af3db1 100644 --- a/libavformat/mxfenc.c +++ b/libavformat/mxfenc.c @@ -79,6 +79,7 @@ typedef struct MXFStreamContext { int interlaced; /// whether picture is interlaced int field_dominance; /// tff=1, bff=2 int component_depth; +int color_siting; int h_chroma_sub_sample; int temporal_reordering; AVRational aspect_ratio; /// display aspect ratio @@ -416,6 +417,7 @@ static const MXFLocalTagPair mxf_local_tag_batch[] = { // CDCI Picture Essence Descriptor { 0x3301, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x02,0x04,0x01,0x05,0x03,0x0A,0x00,0x00,0x00}}, /* Component Depth */ { 0x3302, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x01,0x04,0x01,0x05,0x01,0x05,0x00,0x00,0x00}}, /* Horizontal Subsampling */ +{ 0x3303, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x01,0x04,0x01,0x05,0x01,0x06,0x00,0x00,0x00}}, /* Color Siting */ // Generic Sound Essence Descriptor { 0x3D02, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x04,0x04,0x02,0x03,0x01,0x04,0x00,0x00,0x00}}, /* Locked/Unlocked */ { 0x3D03, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x05,0x04,0x02,0x03,0x01,0x01,0x01,0x00,0x00}}, /* Audio sampling rate */ @@ -996,6 +998,8 @@ static void mxf_write_cdci_common(AVFormatContext *s, AVStream *st, const UID ke unsigned desc_size = size+8+8+8+8+8+8+8+5+16+sc-interlaced*4+12+20; if (sc-interlaced sc-field_dominance) desc_size += 5; +if (sc-color_siting != 0xFF) +desc_size += 5; mxf_write_generic_desc(s, st, key, desc_size); @@ -1030,6 +1034,12 @@ static void mxf_write_cdci_common(AVFormatContext *s, AVStream *st, const UID ke mxf_write_local_tag(pb, 4, 0x3302); avio_wb32(pb, sc-h_chroma_sub_sample); +if (sc-color_siting != 0xFF) { +// color siting +mxf_write_local_tag(pb, 1, 0x3303); +avio_w8(pb, sc-color_siting); +} + If the value as far as we can determine is unknown, should we not write that value (0xFF), rather than leave the metadata out, and hope that any reader will assume the 2009 default rather than the previous, different default? It would seem to me to be more universal. I'm generally wary of leaving out values just because they are some default especially if that default is subject to change. // frame layout mxf_write_local_tag(pb, 1, 0x320C); avio_w8(pb, sc-interlaced); @@ -2037,11 +2047,29 @@ static int mxf_write_header(AVFormatContext *s) // Default component depth to 8 sc-component_depth = 8; sc-h_chroma_sub_sample = 2; +sc-color_siting = 0xFF; if (pix_desc) { sc-component_depth = pix_desc-comp[0].depth_minus1 + 1; sc-h_chroma_sub_sample = 1 pix_desc-log2_chroma_w; } +switch (st-codec-chroma_sample_location) { +case AVCHROMA_LOC_TOPLEFT: sc-color_siting = 0; break; +case AVCHROMA_LOC_LEFT:sc-color_siting = 6; break; +case AVCHROMA_LOC_TOP: sc-color_siting = 1; break; +case AVCHROMA_LOC_CENTER: sc-color_siting = 3; break; If these mappings are correct, then the AVCHROMA_LOC_ names are certainly not intuitive (and the comments unhelpful), and bear little relation to the way SMPTE S377 describes the positioning! e.g. pixfmt.h line 541- AVCHROMA_LOC_TOPLEFT = 3, /// DV SMPTE S377 page 165- 00h coSiting...as in ITU-R Rec 601, SMPTE 314M 4:1:1 or MPEG-2 4:2:2. +case AVCHROMA_LOC_UNSPECIFIED: +if (pix_desc) { +if (pix_desc-log2_chroma_h == 0) { +sc-color_siting = 0; +} else if (pix_desc-log2_chroma_w == 1 pix_desc-log2_chroma_h == 1) { +switch (st-codec-codec_id) { +case AV_CODEC_ID_MPEG2VIDEO: sc-color_siting = 6; break; Only true for 4:2:0 mpeg2, what about 4:2:2 (XDCAM-HD) or does the 'if' filter that out? (I'm not that familiar with the pix_desc struct). +} +} +} +break; +} mxf-timecode_base = (tbc.den + tbc.num/2) / tbc.num; spf =
Re: [FFmpeg-devel] [PATCH] avformat/mxfenc: Add color siting element
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 19/05/15 14:11, Michael Niedermayer wrote: On Tue, May 19, 2015 at 12:07:24PM +0100, tim nicholson wrote: On 19/05/15 01:33, Michael Niedermayer wrote: The default is assumed to be 0xFF, which is what the 2009 spec lists , the older version i have lists 0 as the default. Signed-off-by: Michael Niedermayer michae...@gmx.at --- libavformat/mxfenc.c| 28 + tests/ref/lavf/mxf | 12 +-- tests/ref/lavf/mxf_d10 |2 +- tests/ref/lavf/mxf_opatom |2 +- tests/ref/lavf/mxf_opatom_audio |2 +- tests/ref/seek/lavf-mxf | 44 +++- - --- 6 files changed, 59 insertions(+), 31 deletions(-) diff --git a/libavformat/mxfenc.c b/libavformat/mxfenc.c index 659c34f..0af3db1 100644 --- a/libavformat/mxfenc.c +++ b/libavformat/mxfenc.c @@ -79,6 +79,7 @@ typedef struct MXFStreamContext { int interlaced; /// whether picture is interlaced int field_dominance; /// tff=1, bff=2 int component_depth; +int color_siting; int h_chroma_sub_sample; int temporal_reordering; AVRational aspect_ratio; /// display aspect ratio @@ -416,6 +417,7 @@ static const MXFLocalTagPair mxf_local_tag_batch [] = { // CDCI Picture Essence Descriptor { 0x3301, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x02,0x04,0x01,0x 05,0x03,0x0A,0x00,0x00,0x00}}, /* Component Depth */ { 0x3302, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x01,0x04,0x01,0x 05,0x01,0x05,0x00,0x00,0x00}}, /* Horizontal Subsampling */ +{ 0x3303, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x01,0x04,0x01,0x 05,0x01,0x06,0x00,0x00,0x00}}, /* Color Siting */ // Generic Sound Essence Descriptor { 0x3D02, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x04,0x04,0x02,0x 03,0x01,0x04,0x00,0x00,0x00}}, /* Locked/Unlocked */ { 0x3D03, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x05,0x04,0x02,0x 03,0x01,0x01,0x01,0x00,0x00}}, /* Audio sampling rate */ @@ -996,6 +998,8 @@ static void mxf_write_cdci_common(AVFormatContex t *s, AVStream *st, const UID ke unsigned desc_size = size+8+8+8+8+8+8+8+5+16+sc-interlaced*4+1 2+20; if (sc-interlaced sc-field_dominance) desc_size += 5; +if (sc-color_siting != 0xFF) +desc_size += 5; mxf_write_generic_desc(s, st, key, desc_size); @@ -1030,6 +1034,12 @@ static void mxf_write_cdci_common(AVFormatCon text *s, AVStream *st, const UID ke mxf_write_local_tag(pb, 4, 0x3302); avio_wb32(pb, sc-h_chroma_sub_sample); +if (sc-color_siting != 0xFF) { +// color siting +mxf_write_local_tag(pb, 1, 0x3303); +avio_w8(pb, sc-color_siting); +} + If the value as far as we can determine is unknown, should we not write that value (0xFF), rather than leave the metadata out, and hope that any reader will assume the 2009 default rather than the previous , different default? It would seem to me to be more universal. I'm generally wary of leaving out values just because they are some default especially if that default is subject to change. i was trying to avoid breaking things by favoring not writing it as is done in git currently. But maybe iam too carefull here, do you think we can safely write 0xFF always ? (which could happen if the user does not provide any information about the chroma location or pixel format I should say so, but I wonder if Tomas has a view. or should this be tested with some applications? if so with what applications ? Currently bmxlib reports an ffmpeg.mxf as:- color_siting: Unknown (value='255') i.e, in the absence of the metadata entry it assumes the correct default, so forcibly setting it will make no difference here (or with any other up to date reader). Its only for anything conforming to the old standard of which I do not know a sample. // frame layout mxf_write_local_tag(pb, 1, 0x320C); avio_w8(pb, sc-interlaced); @@ -2037,11 +2047,29 @@ static int mxf_write_header(AVFormatContext *s) // Default component depth to 8 sc-component_depth = 8; sc-h_chroma_sub_sample = 2; +sc-color_siting = 0xFF; if (pix_desc) { sc-component_depth = pix_desc-comp[0].depth_m inus1 + 1; sc-h_chroma_sub_sample = 1 pix_desc-log2_chrom a_w; } +switch (st-codec-chroma_sample_location) { +case AVCHROMA_LOC_TOPLEFT: sc-color_siting = 0; break; +case AVCHROMA_LOC_LEFT:sc-color_siting = 6; break; +case AVCHROMA_LOC_TOP: sc-color_siting = 1; break; +case AVCHROMA_LOC_CENTER: sc-color_siting = 3; break; If these mappings are correct, then the AVCHROMA_LOC_ names are certainly not intuitive (and the comments unhelpful), and bear little relation to the way SMPTE S377 describes the positioning
Re: [FFmpeg-devel] [PATCH 1/2] avformat/mxfenc: Set the component depth from the pixel format if available
On 16/05/15 19:18, Michael Niedermayer wrote: Signed-off-by: Michael Niedermayer michae...@gmx.at --- libavformat/mxfenc.c |7 +++ 1 file changed, 7 insertions(+) diff --git a/libavformat/mxfenc.c b/libavformat/mxfenc.c index 5966792..14449c9 100644 --- a/libavformat/mxfenc.c +++ b/libavformat/mxfenc.c @@ -39,6 +39,7 @@ #include libavutil/random_seed.h #include libavutil/timecode.h #include libavutil/avassert.h +#include libavutil/pixdesc.h #include libavutil/time_internal.h #include libavcodec/bytestream.h #include libavcodec/dnxhddata.h @@ -2029,10 +2030,16 @@ static int mxf_write_header(AVFormatContext *s) } if (st-codec-codec_type == AVMEDIA_TYPE_VIDEO) { +const AVPixFmtDescriptor *pix_desc = av_pix_fmt_desc_get(st-codec-pix_fmt); // TODO: should be avg_frame_rate AVRational rate, tbc = st-time_base; // Default component depth to 8 sc-component_depth = 8; + +if (pix_desc) { +sc-component_depth = pix_desc-comp[0].depth_minus1 + 1; +} + mxf-timecode_base = (tbc.den + tbc.num/2) / tbc.num; spf = ff_mxf_get_samples_per_frame(s, tbc); if (!spf) { LGTM -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 2/2] avformat/mxfenc: Set horizontal chroma subsample value from pixel format if available
On 16/05/15 19:18, Michael Niedermayer wrote: Signed-off-by: Michael Niedermayer michae...@gmx.at --- libavformat/mxfenc.c |7 +-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/libavformat/mxfenc.c b/libavformat/mxfenc.c index 14449c9..659c34f 100644 --- a/libavformat/mxfenc.c +++ b/libavformat/mxfenc.c @@ -79,6 +79,7 @@ typedef struct MXFStreamContext { int interlaced; /// whether picture is interlaced int field_dominance; /// tff=1, bff=2 int component_depth; +int h_chroma_sub_sample; int temporal_reordering; AVRational aspect_ratio; /// display aspect ratio int closed_gop; /// gop is closed, used in mpeg-2 frame parsing @@ -1027,7 +1028,7 @@ static void mxf_write_cdci_common(AVFormatContext *s, AVStream *st, const UID ke // horizontal subsampling mxf_write_local_tag(pb, 4, 0x3302); -avio_wb32(pb, 2); +avio_wb32(pb, sc-h_chroma_sub_sample); // frame layout mxf_write_local_tag(pb, 1, 0x320C); @@ -2035,9 +2036,11 @@ static int mxf_write_header(AVFormatContext *s) AVRational rate, tbc = st-time_base; // Default component depth to 8 sc-component_depth = 8; +sc-h_chroma_sub_sample = 2; if (pix_desc) { -sc-component_depth = pix_desc-comp[0].depth_minus1 + 1; +sc-component_depth = pix_desc-comp[0].depth_minus1 + 1; +sc-h_chroma_sub_sample = 1 pix_desc-log2_chroma_w; } mxf-timecode_base = (tbc.den + tbc.num/2) / tbc.num; LGTM -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH]Allow easy png streamcopying
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 27/04/15 13:16, Nicolas George wrote: L'octidi 8 floréal, an CCXXIII, Michael Niedermayer a écrit : the code changes look ok The phrasing seems to invite this: The commit message does not shock me. Seriously, I do not know how my message about commit messages ended be ing so long, I do not intend to become a commit message tyrant, I just wanted to raise awareness about the issue. You made some sensible points, that will hopefully lead to clearer future commits, if it took that many words to make the point so be it. I do not consider you a tyrant [ yet ;) ]. [..] - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.22 (GNU/Linux) iQEcBAEBAgAGBQJVQdz+AAoJEAwL/ESLC/yD2ggH/jDHgzJohwr/MFrqpNtSmBCZ qs3x2as/k9UjxU2yINv/q8MYJaM3gl1Gnr6EuZ/apTxggEXYI/DW4yf1smeKaTW0 pu4eAQZSKU8xJ3jCGGzcokR27ZVlNdYg32SkCh7mw1DQ9UlXG+fXL5uZVP10TPGz TpgHXHjZ3KoYxbXU1MScrxU3qoQO0oV3PLRr647U1DjjNn8WZQkqbJwEdsTSL//q S/UYuELzWXkLwumYW4VdoczEwnjchXQypGv8CwuhmBnvNnMRRuJm3WgZL5lKgI0c RkVK8PFSUP1Exa0Q/OGHoEjTxLrk/B3Z+6XHq/9yBkWb5dfIjBtJYSSUJQE9oA0= =MtTV -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] Relaxed pattern to find ProRes in MXF.
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 01/04/15 16:32, Steve Dierker wrote: Hello, I support this patch if you didn't make the file yourself. I only had one sample. This sample is from the same customer as the other ProRes in MXF sampl e I uploaded for the bug. (Doner Advertisment) I found it while searching for the reason why FFmpeg refused to encode some files after the bug was fixed. I don't know if the pattern is to relaxed or if there are any other ul's indicating ProRes. I just orientated myself at the DnxHD definitions. DNX is not in the private UL scope, and has three ranges in ffmpeg, one of which I believe to be completely wrong (SMPTE RP224 lists them as tiff formats), but was obviously included for some reason at some stage, so I'm not sure its a good template. It would be good to get a full set of Avid Prores.mxf samples to try and see how they have structured their UL's, I only have windows machines currently to hand, so that is a way off. I am currently looking at the whole range of UL's to see if they can be tidied up, but that will also take some time, so perhaps in the interests of progress use your suggestion for now *but* change the inline comment to something like:- /* Avid MC7 Prores */ That will assist in identifying why the entry is there if we need to update things, especially if the codec moves into the officially recognised scope, as DNX did. Given the interest in Prores in MXF from AMWA that is quite likely imho. best, Steve [...] - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.22 (GNU/Linux) iQEcBAEBAgAGBQJVHQlxAAoJEAwL/ESLC/yDIfYIAMKTctzbP4qNVWbnOLvlBz/g I9JNwvYIR33Jf18bsDJJMX7QaPgnO8y3yExAOHZM7ZCEICK130b79nTWQpVrLhB/ GChnExo69UO7xSprr+RG1hypxnFw6RTf9kK2IpT1aFxRDDSXbiYoG4JyQWMLdo7w O+Mo1/Hz+ftUkOO7twGK4rOwMLl+LjyLhAOXwQ9/xKf57g3cjOB4wBIIk78apOLT zTD9eD2+xH/L8TciLsXC5cGPbU9UJT1JukVuKbF/7P4edLNKRno4RunJjK29poIv iDBSWYd6Js5f3oDZZlyKCnElwPkYEjPUN7h8Zx0tf3M9/mbgjCI3+K0lNBbuEAk= =97j1 -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] Relaxed pattern to find ProRes in MXF.
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 02/04/15 10:57, Steve Dierker wrote: Hello, I am currently looking at the whole range of UL's to see if they can be tidied up, but that will also take some time, so perhaps in the interests of progress use your suggestion for now *but* change the inline comment to something like:- /* Avid MC7 Prores */ That will assist in identifying why the entry is there if we need to update things, especially if the codec moves into the officially recognised scope, as DNX did. I changed the comment and updated the patch. Here is the new file. +{ { 0x06,0x0E,0x2B,0x34,0x04,0x01,0x01,0x02,0x0d,0x01,0x03,0x01,0x02,0x04,0x 60,0x02 }, 14, AV_CODEC_ID_MPEG2VIDEO }, /* Ikegami */ Errm where did this coming from, its unrelated (and wrong afaik as its a container UL). btw: Is this the intended way to offer patches to FFmpeg? if you mean git format-patch against master, yes. best, Steve [...] - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.22 (GNU/Linux) iQEcBAEBAgAGBQJVHS0dAAoJEAwL/ESLC/yDEhYIALvaJcXst47BXby0UbhQzpLy HVGOHUQzvDIHgcbUq+K7o3cs9JCrWvBKZCX3Wqvi9rPsB7scQqucTv0bTiz+vjO6 Y5QQCwRIitExcefOsDfZzy62szqg/8oUSQsohkJOMxdMt6904ucwVQFdzy7RJBVU khgfCffoCCMeQlBZr7Lm+48HpioGwZvZXLIT9aJdzct886gqaS1MpCEsVDStsXTs GH+mHp+64euNekpnfhMZ0xUl4vxKAOynR6tvfgCq5big76vP+aiVmKE7ovygep7I YROkUlgkBWb+rphTZ7XXKLPJni6KC3OHML3IHY5FirQhmuPp2hBzS+AMmgLuSlQ= =+Y3d -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] Relaxed pattern to find ProRes in MXF.
On 31/03/15 13:41, Steve Dierker wrote: Hello List, I found another MXF File containing ProRes with the following codec_uls: 060E2B34040101010E04020102110500 Therefor I relaxed the pattern. Are the Prores variants likely to be that big? Difficult I know for private UL's I wonder if we shouldn't be a little more cautous in widening the scope, where was the sample from? Related to issue #4349 --- libavformat/mxf.c |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) [...] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH] Add 'Presentation Y offset' metadata
Some software mis decodes IMX material in mxf if this metadata field is missing. See the discussion on the ffmpeg ML:- [FFmpeg-user] How to set 3 specific metadata flags (ITU601/displayoffset) in FFmpegs IMX50 MXF-OP1a encoding? This patch add this field to mxf wrapped material. -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 From 7ab1565acb1320a6eaafd47cbae86a8415419274 Mon Sep 17 00:00:00 2001 From: Tim Nicholson tim.nichol...@bbc.co.uk Date: Mon, 30 Mar 2015 11:11:17 +0100 Subject: [PATCH] libavformat/mxfenc.c: Add 'Presentation Y offset' metadata Previously unset, and some software mishandles files if it is absent Signed-off-by: Tim Nicholson tim.nichol...@bbc.co.uk --- libavformat/mxfenc.c | 7 ++- tests/ref/lavf/mxf| 6 +++--- tests/ref/lavf/mxf_d10| 2 +- tests/ref/lavf/mxf_opatom | 2 +- 4 files changed, 11 insertions(+), 6 deletions(-) diff --git a/libavformat/mxfenc.c b/libavformat/mxfenc.c index ac60357..7b400b3 100644 --- a/libavformat/mxfenc.c +++ b/libavformat/mxfenc.c @@ -406,6 +406,7 @@ static const MXFLocalTagPair mxf_local_tag_batch[] = { { 0x3202, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x01,0x04,0x01,0x05,0x02,0x01,0x00,0x00,0x00}}, /* Stored Height */ { 0x3209, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x01,0x04,0x01,0x05,0x01,0x0C,0x00,0x00,0x00}}, /* Display Width */ { 0x3208, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x01,0x04,0x01,0x05,0x01,0x0B,0x00,0x00,0x00}}, /* Display Height */ +{ 0x320B, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x01,0x04,0x01,0x05,0x01,0x0E,0x00,0x00,0x00}}, /* Presentation Y offset */ { 0x320E, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x01,0x04,0x01,0x01,0x01,0x01,0x00,0x00,0x00}}, /* Aspect Ratio */ { 0x3201, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x02,0x04,0x01,0x06,0x01,0x00,0x00,0x00,0x00}}, /* Picture Essence Coding */ { 0x3212, {0x06,0x0E,0x2B,0x34,0x01,0x01,0x01,0x02,0x04,0x01,0x03,0x01,0x06,0x00,0x00,0x00}}, /* Field Dominance (Opt) */ @@ -978,7 +979,7 @@ static void mxf_write_cdci_common(AVFormatContext *s, AVStream *st, const UID ke int stored_height = (st-codec-height+15)/16*16; int display_height; int f1, f2; -unsigned desc_size = size+8+8+8+8+8+8+5+16+sc-interlaced*4+12+20; +unsigned desc_size = size+8+8+8+8+8+8+8+5+16+sc-interlaced*4+12+20; if (sc-interlaced sc-field_dominance) desc_size += 5; @@ -1003,6 +1004,10 @@ static void mxf_write_cdci_common(AVFormatContext *s, AVStream *st, const UID ke mxf_write_local_tag(pb, 4, 0x3208); avio_wb32(pb, display_heightsc-interlaced); +// presentation Y offset +mxf_write_local_tag(pb, 4, 0x320B); +avio_wb32(pb, (st-codec-height - display_height)sc-interlaced); + // component depth mxf_write_local_tag(pb, 4, 0x3301); avio_wb32(pb, sc-component_depth); diff --git a/tests/ref/lavf/mxf b/tests/ref/lavf/mxf index 8ead434..5fb984b 100644 --- a/tests/ref/lavf/mxf +++ b/tests/ref/lavf/mxf @@ -1,9 +1,9 @@ -306708cc2ad2414def89fa2f3c0bfc5c *./tests/data/lavf/lavf.mxf +613b160bf09927661d9455dd975ad780 *./tests/data/lavf/lavf.mxf 525369 ./tests/data/lavf/lavf.mxf ./tests/data/lavf/lavf.mxf CRC=0xdbfff6f1 -f465084f0c365926a81aab56fb6b945c *./tests/data/lavf/lavf.mxf +64471cfc480751f2ca7998c4edbc7a02 *./tests/data/lavf/lavf.mxf 560697 ./tests/data/lavf/lavf.mxf ./tests/data/lavf/lavf.mxf CRC=0x11a6178e -52fc707e1177c97232e2537168c232e6 *./tests/data/lavf/lavf.mxf +a546bd6d8fe1608690bf2fc326cab0c3 *./tests/data/lavf/lavf.mxf 525369 ./tests/data/lavf/lavf.mxf ./tests/data/lavf/lavf.mxf CRC=0xdbfff6f1 diff --git a/tests/ref/lavf/mxf_d10 b/tests/ref/lavf/mxf_d10 index 71707ca..8201557 100644 --- a/tests/ref/lavf/mxf_d10 +++ b/tests/ref/lavf/mxf_d10 @@ -1,3 +1,3 @@ -8f601d5b55a0665cc105a115dc8b3af0 *./tests/data/lavf/lavf.mxf_d10 +22dca5c8d62fe7ad1c56416ae736d6c1 *./tests/data/lavf/lavf.mxf_d10 5330989 ./tests/data/lavf/lavf.mxf_d10 ./tests/data/lavf/lavf.mxf_d10 CRC=0x6c74d488 diff --git a/tests/ref/lavf/mxf_opatom b/tests/ref/lavf/mxf_opatom index 5529e5b..0dd59c1 100644 --- a/tests/ref/lavf/mxf_opatom +++ b/tests/ref/lavf/mxf_opatom @@ -1,3 +1,3 @@ -0f753a141424e2a1b44e6390f70172eb *./tests/data/lavf/lavf.mxf_opatom +5df0bb1083cbe0ef3a70e4a93e44d812 *./tests/data/lavf/lavf.mxf_opatom 4717113 ./tests/data/lavf/lavf.mxf_opatom ./tests/data/lavf/lavf.mxf_opatom CRC=0xbdd696b9 -- 1.9.1 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] Add 'Presentation Y offset' metadata
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 30/03/15 13:21, Michael Niedermayer wrote: On Mon, Mar 30, 2015 at 12:31:04PM +0200, tomas.har...@codemill.se wro te: On 2015-03-30 12:19, tim nicholson wrote: Some software mis decodes IMX material in mxf if this metadata field is missing. See the discussion on the ffmpeg ML:- [FFmpeg-user] How to set 3 specific metadata flags (ITU601/displayoffset) in FFmpegs IMX50 MXF-OP1a encoding? This patch add this field to mxf wrapped material. Looks fine, but I know from experience that all this stuff gets tricky quickly. Like there should be mechanisms in lavf for dealing with presentation rectangles and all that jazz, including different aspect ratios at different layers. should i apply this patch or wait ? Apply, as per IRC comments.. [...] - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v1 iQEcBAEBAgAGBQJVGUmLAAoJEAwL/ESLC/yDmY0H/06658q5T2xtRpu2SRDlyLAY GVLbxpnlWu6s4ss4uB3GRmuGvlZuj2SmJxsvn4s7lKDRiEQIByw2FWIuws5y8CMQ +KhN9nXD2V5D+bHIBTeodb4vfFwLtA33ImPqrpnRawjvQi+b+Bm8Y1qUP1pRxEDF yKOVH1nG2ynNHBbzZgxNAhboGk2nIBHO9JRyMeGSR8XMCiQlsiXPvZcgLXVQ0htR e8DORhEHyNBKX6X6TxJlbqMwakPb+p9SfMXkjKdgzsrWyjz76gHqesDhgnhGwJ8S ApM+ja8Xiy3Q9npNmPxu2RpduDXJ6pJgt87rSn9ZVBs/c2SaT7/pROLL/Pj33T4= =VFpd -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] Add 'Presentation Y offset' metadata
On 30/03/15 11:31, tomas.har...@codemill.se wrote: On 2015-03-30 12:19, tim nicholson wrote: Some software mis decodes IMX material in mxf if this metadata field is missing. See the discussion on the ffmpeg ML:- [FFmpeg-user] How to set 3 specific metadata flags (ITU601/displayoffset) in FFmpegs IMX50 MXF-OP1a encoding? This patch add this field to mxf wrapped material. Looks fine, but I know from experience that all this stuff gets tricky quickly. Like there should be mechanisms in lavf for dealing with presentation rectangles and all that jazz, including different aspect ratios at different layers. Well I thought about adding the X offset too for completeness, but currently due to the lack of the above mechanisms there didn't seem much point as there was nothing to set it from.. (one use would be the 720/702 blanking which some systems also want, but I can see no current programmatical way of setting it.) /Tomas [..] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] Why is writing the colr atom not the default in the mov muxer?
On 21/02/15 01:34, Dave Rice wrote: Hi Robert, Kevin, On Feb 20, 2015, at 9:56 AM, Robert Krüger krue...@lesspain.de wrote: Am Freitag, 20. Februar 2015 schrieb Kevin Wheatley : On Fri, Feb 20, 2015 at 1:30 PM, Robert Krüger krue...@lesspain.de javascript:; wrote: if I read the code correctly, the colr atom is only written in the mov muxer if the flag write_colr is specified. Was that behaviour chosen to have better backward compatibility or is there another reason not to write this standard atom by default? I chose that way to preserve the older behaviour, as it can change how files will be interpreted. I assumed that but isn't the change then a change for the better (then maybe requiring a version bump and an entry in the release notes)? After all Apple muxers write it by default as well and not trusting the input metadata seems to me like something that should be opt-out rather than opt-in but that's just my 2c. That's also my two cents and I also wondered why users have to opt-in to a correctly written file. The QuickTime spec says that colr is required with some streams (such as raw uyvy422 and v210), see: https://developer.apple.com/library/mac/technotes/tn2162/_index.html#//apple_ref/doc/uid/DTS40013070-CH1-TNTAG9. I'd propose that colr be written by default. The interpretation may be different but, since a file with a colr atom is more self-descriptive, the interpretation is more likely to be correct. +1 Dave Rice [...] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] nvenc: Compensate for hardware trying to mess with aspect ratio of DVD content.
-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 On 17/01/15 23:52, Nicolas George wrote: L'octidi 28 nivôse, an CCXXIII, Kieran Kunhya a écrit : There is a very simple way of flagging content that is supposed to comply with BT601: the SAR is 512/351. If SAR is 64/45, that means someone before nvenc decided that the video is not expected to conform with BT601, and nvenc has no right to decide otherwise. You fail to realise this but SARs are not necessarily arithmetic for historical reasons. Care to explain how historical reasons can change the laws of geometry? If a visible surface with a physical aspect ratio of 16/9 is cut into 702×576 identical rectangular pixels, then the aspect ratio of each pixel is (16/9) / (702/576) = 512/351. Can you give one good reason for FFmpeg to use, internally, any other value? I am struggling to follow some of the arguments, which seem far to heated to be as useful as they could be. However, as I understand the issue, whilst the above maths sound fine, the visible surface mentioned above is framed within an actual surface of 720x576 pixels which includes the 'digital overscan' Taking the Sample/pixel aspect ratio (SAR) calculated above and feeding that back into calculating a display aspect ratio *for the whole frame* leads to a value that is not 16:9 (or 4:3), and it appears to be this figure which is then reported as the DAR, rather than that of the active picture. This is I think the disjoint between DAR and SAR that Kieran refers to, there is only a mathematical relationship if you include the overscan. It would appear that in some cases the SAR is calculated back from a specified DAR of 16/9 being applied as the aspect ratio of the whole frame and this seems to be the usual cause of issues (Adobe had to admit they had got it wrong for years as they had assumed that 720x576 was 4/3 aspect ratio). I am not sure what is currently wrong, and where, but please let us not 'fix' one scenario with a dirty hack that breaks hoers. Regards, - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2 iQEcBAEBCAAGBQJUvR+WAAoJEAwL/ESLC/yDE2sH/0/ig2fasBb4QdEjx3wkhKSX SXQP09mt44LNY9hP+EtiDQWyt1Wv4K5XuGwMSuyb48TRilxkjzxvej2m7Gh9oIny XYuk+MIOycU/JLrN7jcY/vLbvyFo9zZirsEZyx2dkHpyt8TvHgNnMuwaAX7SBHPf I73icHhzfipPtambgqRtRuxRThQ3yeCTiTWdF24VWywJhjGgT41TyeqpREfRoptb +uqKbvQheWWe8W3cOAzzmud3PXPqI8fj7X/Y56I0qqEWaRLxEcawOQziE9qK7qCs 7256QUgmEI3gJTfrVY7RJ5EFMM512IhKGCGAUNW1clrb9r2y8M4+ZZC1Mgo9UcE= =pjDp -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] Unable to transcode audio from broadcast ts stream
On 08/01/15 06:58, William Juwono wrote: Hi, I am having problem transcoding the audio of the following file. This is captured from broadcast ts stream.http://www.datafilehost.com/d/cb3d03be If I use acodec copy, the resulting audio played fine, however, if I transcode to other formats, ie: aac, mp3, it resulted in chopped audio.Can someone looked at this? ThanksWilliam [..] As this is the developer list, you are unlikely to get an answer to this question. May I suggest reposting to the ffmpeg user list, and in order to save the inevitable first reply, please include complete uncut console output for a failed trasncode. -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] libavformat/mxfdec.c: export the full UMID as metadata
On 06/01/15 19:01, Mark Reid wrote: On Tue, Jan 6, 2015 at 12:03 AM, tim nicholson nichot20-at-yahoo@ffmpeg.org wrote: On 05/01/15 20:41, Mark Reid wrote: Hi, This patch exports the full umid of packages as metadata. ffmpeg currently only exports the material number portion of the umid. The new format is ISO label-length-instance number-material number example: 060a2b340101010101010f00-13-00-53dc416b9a770251060e2b347f7f2a80 I think this is a good idea, but if we are going to make the change can we make the exported format be SMPTE330 compliant? Annex C (normative) states:- By default, the hexadecimal representation should be preceded by the identifier ‘0x’. This default method of identification will result in 66 hexadecimal characters to represent a basic UMID as follows: 0x060A2B340101010501010D13 ... etc. The hexadecimal letters should be encoded as upper case (A...F). sounds like the right thing to do to me. so if I understand correctly 060a2b340101010101010f00-13-00-53dc416b9a770251060e2b347f7f2a80 should be just 0x060A2B340101010101010F00130053DC416B9A770251060E2B347F7F2A80 easy enough, I'll submit a new patch doing that instead. ___ [..] That's right. drop the '-' add the 0x prefix and upper case it. Saves a byte :) -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] libavformat/mxfdec.c: export the full UMID as metadata
On 05/01/15 20:41, Mark Reid wrote: Hi, This patch exports the full umid of packages as metadata. ffmpeg currently only exports the material number portion of the umid. The new format is ISO label-length-instance number-material number example: 060a2b340101010101010f00-13-00-53dc416b9a770251060e2b347f7f2a80 I think this is a good idea, but if we are going to make the change can we make the exported format be SMPTE330 compliant? Annex C (normative) states:- By default, the hexadecimal representation should be preceded by the identifier ‘0x’. This default method of identification will result in 66 hexadecimal characters to represent a basic UMID as follows: 0x060A2B340101010501010D13 ... etc. The hexadecimal letters should be encoded as upper case (A...F). Some applications use the umids to link to mxf media instead using file paths. I've been using ffmpeg for some AAF work and have encountered MXF files with different ISO labels and having just the material number alone is not enough to link to media for Media Composer. For more on umids, this paper explains them pretty well http://www.digitalpreservationeurope.eu/publications/briefs/UMID_Unique%20Material%20Identifier.pdf umids also have a optional extened 32 bytes, but I've yet to see a file that has them. Mark Reid (1): libavformat/mxfdec.c: export the full UMID as metadata libavformat/mxfdec.c | 39 --- 1 file changed, 32 insertions(+), 7 deletions(-) -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] lavfi: add deinterleave filters
On 06/01/15 16:31, compn wrote: On Tue, 6 Jan 2015 10:10:53 + Paul B Mahol one...@gmail.com wrote: On 1/6/15, Stefano Sabatini stefa...@gmail.com wrote: On date Monday 2015-01-05 16:33:44 +, Paul B Mahol encoded: Signed-off-by: Paul B Mahol one...@gmail.com 3 files changed, 147 insertions(+) create mode 100644 libavfilter/f_deinterleave.c select works as this, with the expression: select=n=N:e='mod(n,N)' I forgot that select can do it, so patch dropped. should that select line go into the manual / examples somewhere? I think the last example is a version of that, (assuming the misssing +1 above is a typo, as if e=0 the frame is dropped) -compn [.] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avfilter/vf_idet: Add analyze_interlaced_flag mode
-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 On 05/01/15 16:57, Michael Niedermayer wrote: On Mon, Jan 05, 2015 at 01:00:41PM +, tim nicholson wrote: On 01/01/15 01:56, Michael Niedermayer wrote: This should allow us to insert idet before scale and let scale have interl=-1 as default in that case Good thinking. How would interl=-1 then get set as default for auto inserted filters. Or did I miss something? it maybe could be set when the filter togeter with idet is inserted i havnt implemented that yet though, idet doesnt support every format that scale supports so it cant be done unconditionally Ahh, right. Am I correct in thinking that currently there is no command line way to globally override the default (interl=0) so that any auto inserted filters use the required settings? I see nothing in sws_flags. I am sort of thinking of a kind of partial implementation workaround where a file is probed and the command line auto constructed based on what is found, then provided the format is OK the default can be changed. This would also be useful in all sorts of other ways when auto inserted filters are likely. [...] - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2 iQEcBAEBCAAGBQJUq5BuAAoJEAwL/ESLC/yDK0YIAIa1jjr9i8k9krY5HhNi7L05 YQrjDZMHZ+35D+sMgeX1/4M/tKNT09bhKhY0YFwIf2v5IDNpGCq7NjfOLuoTntSt P/KFlij9J3zjIhs/nqNt/oKBQRF1ZbFYaQlszeNk84DONHT9qNxODgL959tRpJIx sepbuJQfYMJcIYwuggGk3UR04iO/IFdqSABJfuuuQGNehuLPS7m0LnYbZ1Tjtco9 YvDQOJduMiZ96lV6EZDZSb/wxYiyHQWRu2qBcdlkQSpjzOV1qR28DOzcYO+ebDSf Z4phc8j97vuiHrr+YAQSVXG0B3+/avfWIRFq3oh+rFnW0T5WqcgeDYqGh/O5VT4= =gWUO -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avfilter/vf_idet: Add analyze_interlaced_flag mode
On 01/01/15 01:56, Michael Niedermayer wrote: This should allow us to insert idet before scale and let scale have interl=-1 as default in that case Good thinking. How would interl=-1 then get set as default for auto inserted filters. Or did I miss something? Signed-off-by: Michael Niedermayer michae...@gmx.at [..] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] doc/filters: Add ascii graphics to clarify what the currently implemented tinterlace modes do
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 07/12/14 19:47, Michael Niedermayer wrote: On Fri, Dec 05, 2014 at 01:43:09PM +0100, Michael Niedermayer wrote: On Thu, Dec 04, 2014 at 04:57:09PM +, tim nicholson wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 04/12/14 16:09, Clément Bœsch wrote: [..] This representation requires way more efforts to me to understand than the one from Michael... Just shows how different minds think/work ;) As the saying goes, one man's meat is another man's poison. I was going for consistency with current work (which I happened to find clear) and I liked having the extra detail of line numbers. iam fine with either but id like to see one of them or a combination pushed. pushing neither would be worst ... so which solution do people prefer? seems people dont care so applied mine, feel free to replace by something else, iam fine with any way of documenting it as long as it is documented People don't want choice, they just want what they want. I suspect those with the most pertinent view on the docs style generally are users new(ish) to ffmpeg who don't subscribe to this list. My experience generally is that end user documentation is best not written by developers as they cannot approach the subject from the view point of an outsider coming new to a feature. This is a luxury ffmpeg is not in a position to be able to afford so we do what we can, and respond as best we can if users squawk. Lets see if there is any further noise on the subject and if so we can look to tweaking things. In the meantime as you say pushing neither would be worst, and you got 100% of the votes cast so progress has been made. [...] - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2 iQEcBAEBAgAGBQJUhVv7AAoJEAwL/ESLC/yDVlgH/0aTM27YoG8dwZ128z+rLzVQ 1+vJq8X9xjVR0yXemMG5QRkWagj7z1hhRuZMtVVkMEKIRkvxxlENwlgNrDryjRVm GPgorWq0jkTiX0DRsF40hxqFwIavJBujZoXAmi/8B2ouJcGaVCjsKBOkDsN3RfEM 1/pbUKKLCMnU3dl49DtIFOThfztuFmDzscbwnqQf05/+rMHpYeHONwir1gsTnrOU nEFkm8Nc84ROCcD4j5hV+GiIumThrDzYS85C9aIu9xodgrsRk13o1KWqvLvFSsrk i69wa9C2T4GpfD6I3DWTs54oGONH2X3yZeKmvFPzK0pBoehm7deKKlO1DJxvahQ= =mtPR -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] doc/filters: Add ascii graphics to clarify what the currently implemented tinterlace modes do
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 02/12/14 17:13, Michael Niedermayer wrote: On Tue, Dec 02, 2014 at 05:02:49PM +, tim nicholson wrote: On 02/12/14 15:33, Michael Niedermayer wrote: Signed-off-by: Michael Niedermayer michae...@gmx.at --- doc/filters.texi | 131 ++ 1 file changed, 131 insertions(+) diff --git a/doc/filters.texi b/doc/filters.texi index 8c16c7a..0ea3955 100644 --- a/doc/filters.texi +++ b/doc/filters.texi @@ -8537,33 +8537,164 @@ Available values are: @item merge, 0 Move odd frames into the upper field, even into the lower field, generating a double height frame at half frame rate. +@example + -- time +Input: +Frame 1 Frame 2 Frame 3 Frame 4 + +1 2 3 4 +1 2 3 4 +1 2 3 4 +1 2 3 4 + +Output: +1 3 +2 4 +1 3 +2 4 +1 3 +2 4 +1 3 +2 4 [..] I find the ascii graphic style of the interlace filter more understandable how do you suggest above (and the other modes) would be documented in that style ? Sorry for the delayed reply ISP issue... I think the main difference is going left to right, not top to bottom for input to output, this then matches not only interlace but the general filtergraph descriptions. I will try and see if I can give an example if you like. [...] - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2 iQEcBAEBAgAGBQJUgBpMAAoJEAwL/ESLC/yDxkEH/1hWM5A1P3V3cz2XH3pF01/0 64e4A9RRx0BpX0pDHqik9aVNQtL+qSBFYkkQGPn8smTdqWEQXfIe/lWQSXODj95T +/Un1V33cxnPznJnTCfxrZEKuYbEQYehyuasGcgzq/L0CIYamGJmuWZ4feZKQ0Vk ZdCAs/ygX9igc7V+41UG5dQu6n+nhIwAkOAJBiGkeVMKu76YH19fF/G4TO7tYczn GbVgKfu78LK8cZZOWl3w42EO7aF98TeG6q29jQsiDABLNa8JocIVmxaijbaIEJ3R 2Eb5yzCeyZadgOrenRzF9iR0W4JvlXK2Expqr15SISRjVJ4L8AjZqs6nPOsraHY= =KKl4 -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avfilter/tinterlace: merge code with interlace
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 03/12/14 13:08, Michael Niedermayer wrote: On Tue, Dec 02, 2014 at 07:43:47PM +0100, Clément Bœsch wrote: --- After this commit, interlace doesn't behave the same if the input frames are already interlaced. Does anyone wants me to keep this behaviour? probably best if its kept but can be forced enabled/disabled with an option. these interlacing flags on the input can be wrong +1 otherwise LGTM [...] - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2 iQEcBAEBAgAGBQJUgBvNAAoJEAwL/ESLC/yDOmMH/3U73HxpvOsW4QwvfBDQFeD0 dMqyo9NgZFUzsNLptORc9ifRXQrDpxvhkmhPtqpCgpgLBV+E88VyXCEo+QPrq8AY l617h6Wb/FbRHT4opYTfU2DsSgdNKalAZacrckY1dGFNEACQRg04n8jGkbyK5gu/ 1YXFknMic92Vq29uwPrymrmvcGyWtjj2goTL/6ikVbAdFGQ6dks21PR412YUqk6z BcVuWXxXv9HZft2pfE1edcT3FEqiJL/iKZmzq91i6vzkyS5xeHXkLgCNc+P3NfOx pPvKl4lFhRFMh4OTxVT82EqUvrFovgH3bRXXocQeIxuzf+BXzrkVZOSMPjtoG1o= =bJWr -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avfilter/tinterlace: merge code with interlace
On 02/12/14 18:43, Clément Bœsch wrote: --- After this commit, interlace doesn't behave the same if the input frames are already interlaced. Does anyone wants me to keep this behaviour? Also, the PTS adjustments might be different, but I didn't follow the current discussions about it so I can't tell what people want. --- libavfilter/Makefile| 2 +- libavfilter/interlace.h | 58 - libavfilter/tinterlace.h| 5 +- libavfilter/vf_interlace.c | 250 libavfilter/vf_tinterlace.c | 44 +++ libavfilter/x86/Makefile| 2 +- libavfilter/x86/vf_interlace_init.c | 47 --- 7 files changed, 49 insertions(+), 359 deletions(-) delete mode 100644 libavfilter/interlace.h delete mode 100644 libavfilter/vf_interlace.c delete mode 100644 libavfilter/x86/vf_interlace_init.c [..] The docs will need updating to remove the vf_interlace entry and add the lp filter options. Maybe that is the time to merge the ascii art graphics and add Michael's additional mode ones as well? -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avfilter/tinterlace: merge code with interlace
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 04/12/14 09:59, Clément Bœsch wrote: On Thu, Dec 04, 2014 at 09:31:47AM +, tim nicholson wrote: On 02/12/14 18:43, Clément Bœsch wrote: --- After this commit, interlace doesn't behave the same if the input frames are already interlaced. Does anyone wants me to keep this behaviour? Also, the PTS adjustments might be different, but I didn't follow the current discussions about it so I can't tell what people want. --- libavfilter/Makefile| 2 +- libavfilter/interlace.h | 58 - libavfilter/tinterlace.h| 5 +- libavfilter/vf_interlace.c | 250 libavfilter/vf_tinterlace.c | 44 +++ libavfilter/x86/Makefile| 2 +- libavfilter/x86/vf_interlace_init.c | 47 --- 7 files changed, 49 insertions(+), 359 deletions(-) delete mode 100644 libavfilter/interlace.h delete mode 100644 libavfilter/vf_interlace.c delete mode 100644 libavfilter/x86/vf_interlace_init.c [..] The docs will need updating to remove the vf_interlace entry and add the lp filter options. Maybe that is the time to merge the ascii art graphics and add Michael's additional mode ones as well? None of the filter disappear and their options don't change. It's just code refactoring. Sorry misread it only looking at the diff. I thought there was a funtionality merger as well as per the irc discussion. [..] - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2 iQEcBAEBAgAGBQJUgFaLAAoJEAwL/ESLC/yDRGkH/j2ZXp+JcWgfvTIG5+/U5Nrk uDA9Gpn+QQKQHjaF+AHCL8d1KPmXlnKgbha2r9V4EBHiSyZDTNOcio9T0IvY+YDG wbnTuCSV6SaBgm8LvKrvB2B/OC4aMFoP8aYyE10Ghqkd2HHG7jGcctPrlAIdmvpP K6VbHFP+RKuvQLvQPX2SPigcGvUQV566azYguzum82k4CV4+kwugUCDNgGeds8rO ySy/DoeTai0cm5A+PA3ojKMaWkuZt5AN6WFFGcYvamwLDa/WZfK8SlcpvPZoH9y9 9DYizi3PDOfyojpXKoMsi6n3zRpKJ9x7TETA+hUZ8mcJDxnhCvFhYIDeQXrt8LY= =m1ni -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] doc/filters: Add ascii graphics to clarify what the currently implemented tinterlace modes do
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 04/12/14 16:09, Clément Bœsch wrote: [..] This representation requires way more efforts to me to understand than the one from Michael... Just shows how different minds think/work ;) As the saying goes, one man's meat is another man's poison. I was going for consistency with current work (which I happened to find clear) and I liked having the extra detail of line numbers. [...] - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2 iQEcBAEBAgAGBQJUgJJlAAoJEAwL/ESLC/yDmdQH/1BHNlTjS9omO0g28P7+TAXu vpGI7kUxJWFH7txUHhRNf33lHbFFmYdKeb/lIfuiPFXvOtUgeTnmARf8RPIN82e4 7LPFIuN9BFuu78GpyjY/ZEQOUhsT8WK/5c/xWzgJDTo87WsqmQcZe1TuJW945U16 mEZnPfCoYnV4OO/Bu68FJjEMHX4MafIB5VKxhK28/6ymbSJl5v8h9uyFzEZ3TZgF 48bSSYNM4T1KwkJPQsDN+0X5YXeAjA+ZTbB1K5qBQAOZYY4Qj2oCSetl09G6XynG FjeYUxsDCwI97/+x2i+0K6UIBj9Ey/iJRUat1QP4F1FoHQKz5WHpMN5/4XYblU0= =ulRP -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avfilter/vf_tinterlace: Favor using standard timebases for the output
On 01/12/14 22:32, Kieran Kunhya wrote: Sent from my mobile device [...] I would strongly advise all users of these filters generate their own timestamps. Which would require much better documentation of such usage. (e.g. the only references to the use of genpts inn ffmpeg docs are in passing on the trac wiki) [...] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] doc/filters: Add ascii graphics to clarify what the currently implemented tinterlace modes do
On 02/12/14 15:33, Michael Niedermayer wrote: Signed-off-by: Michael Niedermayer michae...@gmx.at --- doc/filters.texi | 131 ++ 1 file changed, 131 insertions(+) diff --git a/doc/filters.texi b/doc/filters.texi index 8c16c7a..0ea3955 100644 --- a/doc/filters.texi +++ b/doc/filters.texi @@ -8537,33 +8537,164 @@ Available values are: @item merge, 0 Move odd frames into the upper field, even into the lower field, generating a double height frame at half frame rate. +@example + -- time +Input: +Frame 1 Frame 2 Frame 3 Frame 4 + +1 2 3 4 +1 2 3 4 +1 2 3 4 +1 2 3 4 + +Output: +1 3 +2 4 +1 3 +2 4 +1 3 +2 4 +1 3 +2 4 [..] I find the ascii graphic style of the interlace filter more understandable -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH]Mention in the documentation that fieldmatch needs cfr input
On 23/10/14 17:06, Dave Rice wrote: On Oct 23, 2014, at 4:05 AM, Clément Bœsch u...@pkh.me wrote: On Tue, Oct 21, 2014 at 09:32:39AM +0200, Carl Eugen Hoyos wrote: Hi! It appears to me that we all know that fieldmatch needs cfr input, but it isn't mentioned in the documentation. Related to ticket #3968. Please comment, Carl Eugen diff --git a/doc/filters.texi b/doc/filters.texi index c70ddf3..bc77623 100644 --- a/doc/filters.texi +++ b/doc/filters.texi @@ -4447,6 +4447,9 @@ and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from which @code{fieldmatch} is based on. While the semantic and usage are very close, some behaviour and options names can differ. +The filter only works for strictly constant frame rate input. If your input +has variable frame rate, try the @ref{pullup} filter. + Well... isn't telecined content supposed to be CFR anyway? I would say it's supposed to be undefined behaviour in any ivtc filter but well. No opinion really. This would be ideal, but some common production applications use VFR. For instance when digitizing with the Log and Capture interface of Final Cut Pro the first and last frame usually have custom durations although most of the rest of the file is 2997/100. There’s another setting in the Final Cut to compensate for dropped frames by extending the duration of the last good captured frame. Not only FCP. Pipelines do it too. Dave Rice _[..] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] Facebook page
On 21/10/14 20:21, Lou Logan wrote: On Tue, 21 Oct 2014 19:30:39 +0200, Thomas Volkert wrote: There was also no response from Facebook within the last days. The behemoth is probably slow to respond. Let's give them some more time then perhaps prod them again after not hearing from them for a week. I'd rather have ffmpeg or nothing. Easy for me to say though since I haven't done a damned thing to help. There currently seem to be 4 'pages' called FFmpeg, one auto generated so that's a bit of work to unpick. [..] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH] avformat/movenc: AVC Intra support
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 28/09/14 02:59, Michael Niedermayer wrote: On Sun, Sep 28, 2014 at 03:53:15AM +0200, Michael Niedermayer wrote: On Sun, Sep 28, 2014 at 02:00:00AM +0100, Kieran Kunhya wrote: On 28 September 2014 00:49, Michael Niedermayer michae...@gmx.at wrote: This allows remuxing AVC intra into mov it does not work with libx264 encoded AVC Intra for unknown reason 1080i should work. 1080p needs some random magic numbers in the bitstream which are probably different for all the framerate/resolution combinations. Blame Apple for having a decoder that violates the H.264 spec - specifically requiring SEIs to have certain information in them. i tried various avcintra 50 and 100 and i didnt find any that worked. They all crashed fcpx (in apples avci decoder in the case i looked at) its of course very possible that i made some systematic mistake if someone has a known to be good set of command line options for x264 to generate a working stream + input + output .h264 streams that might (or might not) help identify what is at fault [...] Some of the DPP AVCI100 test samples were created using x264* so could be used for reference. Or I could knock up something if I can find my reference CL I used last time. *Using libx264 in ffmbc. - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.19 (GNU/Linux) iQEcBAEBAgAGBQJUKm7yAAoJEAwL/ESLC/yD/mgIALrkpV2zGILv5X3TAslK6KMl 2f9czw70hPQ6uBXLUiUf3Dr2Fmg7GQ5gP/rZpPIKNiZPXcrdQwTsrU6k1FNNUUZi OnR9BR3dfwjrH8ekYyy5i3ZjHRktlGBPD3adyQWpENahHtui0jdGMpnMeSc5Mro9 yN3FTRLlB01GHmTR4KG2QO2H6qenMe5jVShAPJbQ2crvJdSI/4mxNpEjLSYBgA5W Fe09SdNx+1xnHuM15Uq7W2yG3N+lmRQZlscB+eWwo1zwb6SpjUYCjcuggnWaBQl8 qxfXuwIMOT0DgVdVVt5gcARZ43cM14e9rZ+eOc9Qq6RJ4/s44Ua3OYfneQurJPw= =yTRs -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH]Do not ask for mxf samples with unknown field dominance
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 10/09/14 21:45, Tomas Härdin wrote: On Tue, 2014-08-19 at 22:30 +0200, Michael Niedermayer wrote: On Tue, Aug 19, 2014 at 01:30:24AM +0200, Carl Eugen Hoyos wrote: Hi! Attached patch removes a request for samples of which we already have several that all work fine. field_dominance can have 256 different values, do we have samples for all ? do they even exist ? Whilst the variable may be able to contain 256 different values in reality there are only two possible states f1 first or f2 first. As far as I recall there are only two (1, 2) and possibly unknown (0). Is there a distinction between unknown and progressive? I don't have the numbers to hand. /Tomas [..] - -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.19 (GNU/Linux) iQEcBAEBAgAGBQJUF+7lAAoJEAwL/ESLC/yD914H/j1iF4mec8zkiVQeySSOOkB+ mebCr5yNhwuYDGQvzCp15nVpLi1y/uyP0TCu+iCJp/pDHR+z8iSO6moKhY+hB0hp 3395I22SujmSa8ODY9NWTrZWqAKnS0Lohfkzva04Tvn31cLskWpQKxPWhBajz0dP d9G4h9O9YhsVBH1L/UhRHRDwDSeghqjTvnw3U63DoGRitFCKZAARATx0O7iH51Xv we3Mk7FAgpWfB5GVQ3VQZWkzydsoQZNjtWnGp4kjnqdONOMY8H6EbiqeQuCY2Tfs yu6OX5KowMAv7YfCrU0J1BlTsl90AA9Rh3KEFhHTjvI75KhAW0KL6x1Oxuz4XcU= =hjO4 -END PGP SIGNATURE- ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [RFC]Revert the default for ffplay to -noautoexit
On 07/09/14 21:39, Carl Eugen Hoyos wrote: Hi! I have no strong opinion on this but I wonder if ffplay isn't primarily a test application for which it is very useful not to exit. +1 Please comment, Carl Eugen [..] -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] Test of mail system pllease ignore
Should munge the from headers -- Tim. Key Fingerprint 38CF DB09 3ED0 F607 8B67 6CED 0C0B FC44 8B0B FC83 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel